Methodology of Political Economy Research

The methodology of political economy research Political economy as an interdisciplinary field of knowledge utilizes a variety of methodological approaches borrowed from economics, political science, sociology, and history. Understanding the methodological toolkit is necessary for the critical analysis of political economy studies and for the development of one's own analytical skills.

Positive and Normative Analysis
A fundamental methodological distinction passes between positive and normative analysis. Positive analysis answers the question “what is” and “why,” seeking to objectively describe and explain observed phenomena. Normative analysis answers the question “what ought to be,” implying value judgments about the desired organization of society. Classical political economy openly combined both approaches: Adam Smith and David Ricardo not only analyzed economic regularities but also put forward recommendations for economic policy. Modern economic science seeks to divide positive and normative, however, political economy recognizes that complete value neutrality is impossible — the very choice of a research question already reflects certain values and priorities.

Quantitative Methods
Quantitative analysis occupies an important place in modern political economy. Econometric methods make it possible to identify causal relationships between political and economic variables, controlling for alternative explanations. Regression analysis is a basic tool for assessing the influence of some variables on others. For example, a researcher can estimate the impact of democracy on economic growth by controlling for education level, geographic location, and historical factors. Panel data combines information about many countries or regions over several time periods. This allows one to account for both cross-national differences and the dynamics of changes within each country. Instrumental variables is a technique for solving the problem of endogeneity, when cause and effect influence each other. A classic example is the use of geographic factors as instruments to estimate the impact of institutions on development (works by Acemoglu, Johnson, Robinson). Difference-in-differences is a method for comparing the changes in a group subjected to an intervention (for example, a reform) with changes in a control group. This allows isolating the effect of a specific policy from general trends. Regression discontinuity is a method for situations where a policy is applied based on a certain threshold (for example, districts where a candidate won by a minimal margin are compared with districts where he lost by a minimal margin).

Qualitative Methods
Quantitative methods have limitations: they show correlations and average effects well, but often do not reveal the mechanisms and context. Qualitative methods complement the picture. Comparative case analysis (case studies) allows for an in-depth examination of specific countries, reforms, or episodes. The method of comparative case study systematically compares several cases, identifying common patterns and important differences. Historical analysis traces the evolution of institutions, policies, and economic structures over time. A historical perspective is necessary to understand path dependence — the dependence of current outcomes on past decisions. Qualitative Comparative Analysis (QCA) is a method based on Boolean algebra that allows one to identify combinations of conditions necessary or sufficient for a particular outcome. QCA is especially useful with a small number of cases, when statistical methods are inapplicable. Process tracing is a method of detailed tracking of cause-and-effect chains within a single case. The researcher collects evidence on how specifically the cause led to the effect, checking alternative explanations.

Modeling in Political Economy
Formal models are mathematical constructs describing the behavior of rational actors under given institutional conditions. Models simplify reality but make it possible to identify the logic of interactions and formulate testable hypotheses. Game theory is the main tool of formal modeling in political economy. Models of strategic interaction analyze how rational actors make decisions, taking into account the expected actions of others. Applications include models of trade negotiations, lobbying, political competition, and collective action. Spatial voting models represent the preferences of voters and candidates as points in a multidimensional space of political positions. The median voter theorem shows that under certain conditions a political outcome is determined by the preferences of the “average” voter. Principal-agent models analyze relations between a “principal” (for example, voters) and an “agent” (a politician), when the agent possesses an informational advantage and may pursue his own interests.

Problems of Causal Identification
A central methodological problem of political economy is the establishment of causal links. Correlation between two variables does not mean causality: the relationship may be reverse (from effect to cause), mediated by a third variable, or random. The problem of endogeneity is especially acute in political economy: policy affects the economy, but the economy also influences policy. Democracies are richer than dictatorships — but is it because democracy fosters growth, or because rich countries are more likely to become democracies? The problem of selection arises when the phenomenon under study (for example, a reform) occurs not by chance but under certain conditions. Countries implementing reforms may systematically differ from those that do not, which complicates the assessment of the reform’s effect itself. The omitted variable problem means that an observed relationship may be explained not by the investigated cause, but by other factors not included in the analysis.

Strategies for Solution
Researchers use various strategies to address causality problems:

Natural experiments — situations where changes in policy or institutions occur for reasons unrelated to the studied outcome. For example, colonial borders drawn with a ruler without regard for local conditions are used to study the impact of artificial division of ethnic groups on conflicts.
Randomized experiments are increasingly used in the political economy of development. Random assignment of programs (for example, conditional cash transfers) among villages makes it possible to accurately assess their effect.
Multiplicity of methods — the use of different approaches to test the same hypothesis. If results are robust under different specifications, data, and methods, this increases confidence in the findings.

Data in Political Economy
The quality of analysis is directly dependent on the quality of data. Political economy uses a variety of sources:

Macroeconomic data — GDP, inflation, trade, investment — collected by national statistical offices and international organizations (IMF, World Bank, OECD).
Political indices measure democracy (Polity, Freedom House, V-Dem), corruption (Transparency International), institutional quality (Worldwide Governance Indicators), economic freedom (Heritage Foundation, Fraser Institute).
Survey data — World Values Survey, Eurobarometer, Afrobarometer — provide information on citizens’ values, attitudes, and behaviors.
Historical data — from population censuses to archival documents — allow for the analysis of long-term processes.

Criticism and Limitations
The methodology of political economy is subject to criticism from various perspectives. Quantitative research is criticized for ignoring context, being mechanistic, and for false precision. Qualitative research is criticized for subjectivity and limited generalizability. Formal models are criticized for unrealistic assumptions. Awareness of the limitations of each method is a sign of methodological maturity. The best studies combine different approaches, using their complementary strengths.