What Comes First: the State, the Market, or Democracy?

One of the oldest questions in political economy is also one of the most practical: in what order do good institutions develop? Do countries first need a capable state? Do they first need markets and economic freedom? Or can democracy arrive before either of these conditions is securely in place?

This is the question taken up by Ryan H. Murphy and Colin O’Reilly in their article “Sequencing institutional development: The modernization hypothesis reconsidered,” published as an article in press in the Journal of Comparative Economics. The paper revisits the modernization hypothesis with modern data and time-series methods. Its conclusion is important because it is both clear and restrained: state capacity seems to precede economic freedom, but there is little evidence that either state capacity or market institutions must precede democracy.

This result matters for development economics, transition economics, and political economy. It also matters for policy debates. If the authors are right, we should be careful about grand narratives saying that countries must become rich before they democratize, or that democratization must wait until the state is already strong. At the same time, the paper suggests that market liberalization may be easier to sustain when the state is already capable of performing its basic functions.

The question: is there a sequence of institutional development?

The modernization hypothesis is usually associated with Seymour Martin Lipset’s classic 1959 argument that economic development makes democracy more likely and more stable. In its simplest form, the idea is intuitive: as countries become richer, more educated, more urbanized, and more socially complex, they become more favorable environments for democratic politics.

Murphy and O’Reilly focus on a broader version of this debate. Instead of asking only whether income comes before democracy, they ask how three institutional dimensions relate to each other over time:

  • State capacity: can the state enforce rules, administer a bureaucracy, control its territory, raise revenue, provide public goods, and uphold the rule of law?
  • Economic freedom: are markets relatively open, property rights protected, money stable, trade freer, and regulation less burdensome?
  • Democracy: are political leaders selected through broad, competitive, and meaningful electoral processes?

The sequencing question can be written as a simple chain:

Capable state → market institutions and prosperity → democracy

This chain combines two familiar ideas. The first is the “state-first” view, often associated with authors such as Samuel Huntington, Francis Fukuyama, and Fareed Zakaria: before countries can liberalize economically or democratize politically, they need a functioning state. The second is the Lipset view: economic development makes democracy more likely to work and to persist.

The value of the paper is that it tests this sequence directly, rather than simply assuming it.

A pedagogical detour: what is state capacity?

“State capacity” can be a confusing term because it is sometimes mistaken for “big government.” That is not what it means here. A state can be large but ineffective, or relatively small but capable. In this paper, state capacity refers to the ability of the state to perform core functions in a predictable and competent way.

Think of state capacity as the institutional infrastructure behind public authority. A capable state can collect revenue without relying only on arbitrary extraction, maintain control over its territory, administer public policy through a reasonably impartial bureaucracy, provide basic public goods, and sustain a legal order. These are not minor details. Markets require contracts, property rights, and credible enforcement. Democracy requires elections, security, and a public administration capable of implementing decisions.

Murphy and O’Reilly use a broad historical measure of state capacity developed by O’Reilly and Murphy. This is an important contribution because many governance indicators begin only in the late twentieth or early twenty-first century. The longer historical panel allows the authors to study institutional dynamics over several decades rather than only in the post-2000 period.

The data: three institutional variables over time

The paper’s main analysis uses a balanced panel of 99 countries, observed every five years from 1975 to 2020. This is well suited to the study of institutions because institutions usually change slowly. Annual changes may be too noisy or too small, while five-year intervals better capture meaningful institutional shifts.

The authors use three main measures:

Concept Measure used in the paper Interpretation Source
State capacity O’Reilly and Murphy’s broad state capacity index The state’s ability to provide public goods, administer policy, control territory, raise revenue, and enforce law State Capacity Index data and code archive; O’Reilly and Murphy (2022)
Democracy V-Dem’s electoral democracy / polyarchy index The extent of electoral competition, broad suffrage, freedom of expression, and meaningful elections V-Dem v14 dataset; archive; Codebook v14
Economic freedom Economic Freedom of the World index The institutional environment for markets: property rights, sound money, trade openness, regulation, and size of government EFW dataset; 2023 report

The last column gives direct links to the data sources and documentation, so readers can inspect the underlying measures rather than treating the indices as black boxes.

One subtle point is worth emphasizing. The paper does not use GDP per capita as the main economic variable. Instead, it uses economic freedom. This means the authors are not only asking whether richer countries become democratic. They are asking whether more market-oriented economic institutions tend to appear before democracy, after democracy, or after improvements in state capacity.

The method: why use panel VAR?

The paper uses panel vector autoregression, or PVAR. The expression sounds technical, but the basic intuition is accessible.

Imagine that we observe three variables for many countries over time: state capacity, economic freedom, and democracy. Each variable may influence the others, and each variable may also be influenced by its own past. A PVAR model allows all three variables to move together dynamically. It does not force us to say from the beginning that one variable is always the cause and the others are always the effects.

The authors use two main tools within this framework.

First, they use Granger causality tests. Despite the name, Granger causality is not causality in a philosophical or experimental sense. It asks a narrower question: do past values of one variable help predict future values of another variable? For example, if past state capacity helps predict future economic freedom, then state capacity “Granger-causes” economic freedom.

Second, they use impulse response functions. These ask: if there is an unexpected improvement in one institutional variable, what happens to the others over the following periods? Since the data are observed in five-year intervals, the paper traces responses over roughly twenty years.

This is a useful strategy because sequencing is fundamentally a temporal question. If we want to know what comes first, we need methods that take time seriously.

The main finding: state capacity tends to come before economic freedom

The most robust result is that state capacity precedes economic freedom. In the Granger causality tests, state capacity helps predict future economic freedom. The reverse is not robustly supported: economic freedom does not clearly predict future state capacity.

The impulse response functions tell a similar story. A positive shock to state capacity is followed by a substantial increase in economic freedom. In the authors’ baseline specification, a one-standard-deviation innovation in state capacity is associated with an increase in economic freedom of more than one standard deviation after fifteen years. They translate this into an economically meaningful change: around 1.6 points on the Economic Freedom of the World scale.

This is not a small result. It suggests that countries do not simply liberalize first and then build competent states afterward. More often, the improvement of state institutions appears to create conditions under which market institutions can develop.

Why would state capacity support market liberalization?

This finding is intuitive once we think about what markets need in order to function. Market liberalization is not only a matter of removing controls. It also requires credible rules. Property rights need courts. Trade needs customs administration. Firms need predictable regulation. Monetary stability requires institutions. Contracts require enforcement.

In weak states, liberalization may create opportunities for rent seeking, monopolization, asset stripping, or selective enforcement. In more capable states, liberalization can be embedded in a rule-based environment. This is why the finding is not simply “more state” versus “more market.” The relevant distinction is between arbitrary state intervention and capable state administration.

The paper’s conclusion echoes a lesson from the post-socialist transitions. The problem was not only whether assets were privatized, but whether privatization occurred in an environment with secure property rights and the rule of law. A market economy is not created by privatization alone; it requires a legal and administrative order that makes private property meaningful.

The non-finding: democracy does not seem to require prior markets or prior state capacity

The second major result is more surprising for modernization theory. The paper finds little support for the claim that economic freedom must precede democracy. It also finds little support for the claim that state capacity must precede democracy.

This does not mean that markets, wealth, or state capacity are irrelevant for democratic survival in every context. The authors are careful about this. The PVAR framework is better suited to studying whether improvements in one variable are followed by improvements in another. It may not fully capture the more specific claim that higher income helps democracies avoid backsliding once democracy already exists.

Still, the paper’s results challenge a strong version of modernization theory. They do not show that countries must first become economically liberal or administratively strong before democratizing. In this respect, the paper pushes back against overly deterministic views of political development.

What the figures show

The paper’s figures are especially helpful for understanding the logic of the results. The impulse response figure on page 10 shows that when state capacity receives a positive shock, economic freedom rises over subsequent periods. The response of democracy is much smaller. The robustness figures in the appendix repeat the exercise under alternative assumptions about the ordering of the variables. The state-capacity-to-economic-freedom relationship remains the most robust pattern.

The country illustrations for Spain and Taiwan also help. In both cases, improvements in state capacity are followed by, or move closely with, later improvements in economic freedom. These examples should not be read as proof by themselves, but they make the mechanism easier to see. A state consolidates its administrative and legal capacity, and market institutions subsequently improve.

How the paper handles robustness

A strong part of the paper is its attention to robustness. The authors test whether the main result depends on particular modeling choices. They vary the number of lags, extend the sample back to 1970 in an alternative specification, test all possible orderings among the three institutional variables, and use an alternative state capacity measure that removes variables overlapping with the economic freedom index.

This last point is important. Since both state capacity and economic freedom partly involve legal institutions, one might worry that the relationship is mechanical: perhaps the same underlying rule-of-law measures appear in both variables. The authors address this by reconstructing the state capacity index without the potentially overlapping components. The main finding remains.

They also estimate specifications with exogenous controls for coups and major oil discoveries. These additions are meant to account for political shocks and resource-curse dynamics that might otherwise complicate the interpretation. The conclusion is unchanged: state capacity robustly precedes economic freedom, while the evidence that economic freedom precedes democracy remains weak.

What should we take away?

The paper gives us a useful way to update the institutional sequencing debate. It does not say that sequencing is irrelevant. But it does say that the only robust sequence in the data is narrower than many theories suggest.

Here are the main lessons.

  1. State capacity and market institutions are complements. A capable state can make economic liberalization more credible and effective.
  2. Economic freedom does not clearly precede democracy. The data do not support a strong version of the idea that countries must liberalize economically before they democratize.
  3. State capacity does not clearly precede democracy either. The evidence does not justify a simple “build the state first, democratize later” rule.
  4. Grand sequences should be treated cautiously. Institutional development is path-dependent, historically specific, and shaped by many small and large shocks.

This is why the paper is valuable. It does not replace one rigid sequence with another. It narrows the claim: state capacity seems to help economic freedom, but the road to democracy is less sequential than modernization theory often implies.

Policy implications: capable states, not necessarily bigger states

For development policy, the paper points toward a balanced institutional view. Market reforms are unlikely to succeed if they are introduced into a setting where rules are unpredictable, courts are weak, tax systems are arbitrary, and public administration is ineffective. A country may formally liberalize, but without state capacity, liberalization may be captured by insiders or undermined by insecurity.

At the same time, the paper does not support delaying democracy until some technocratic threshold has been reached. There is no clear evidence in the paper that democracy must wait for either market institutions or strong state institutions. This is an important corrective to arguments that treat democracy as a luxury good available only after development has occurred.

The more reasonable lesson is that state-building, market-building, and democracy-building interact in complex ways. But among these relationships, the most robust is the connection from state capacity to economic freedom.

A final thought

The most interesting papers are often those that complicate a familiar story. Murphy and O’Reilly do exactly that. Modernization theory is not simply confirmed or rejected. Instead, it is decomposed.

Yes, there is evidence for a state-first mechanism in the development of market institutions. Better state capacity tends to come before greater economic freedom. But no, the evidence does not show that markets must come before democracy, nor that democracy must wait for a strong state.

The result is a more modest and more useful theory of institutional development: build capable states because they help markets work, but do not assume that democracy has a single institutional prerequisite.

Sources, data, papers, and replication links

The links below collect the main article, replication-related archive links, the source datasets, and selected papers mentioned in the discussion.

Main article and replication

Data sources

Selected papers and methodological sources

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Related Posts

France’s Fiscal Weakness, Italy’s External Constraint: Two Paths in the Euro Area

Why France and Italy? The new research published in the Journal of International Money and Finance “What are the causal links between fiscal and external sustainability in the…

The Political Economy of Geoeconomics

This post builds on Jeffry Frieden’s comment on geoeconomics in the NBER Macroeconomics Annual. Frieden’s central argument is that geoeconomic policy should not be analyzed as if states…