Skip to main content

Effects of Structural Change on Income Inequality

Periodic Reporting for period 2 - STRCHANGEINEQ (Effects of Structural Change on Income Inequality)

Reporting period: 2018-08-01 to 2019-07-31

This project aims to identify the distributional effects of the structural change phenomenon. Structural change can be defined as the reallocation of economic activity across the sectors (mainly manifested as a decline in the agriculture, a rise and a decline in manufacturing, and a rise in services) during long run development. The project develops a dynamic macroeconomic model that introduces heterogeneities across workers and asset owners into the structural change models. With this methodology, we are able to characterize not only the aggregate patterns of structural change, but also its distributional consequences. By uncovering the long term factors behind the evolution of income distribution during structural change, the proposed model is highly relevant to the most recent debates of rising income inequality in developed countries. The theoretical model is calibrated to assess and quantify the historical roots of the current distributional picture in U.S. and Europe. In doing so, it determines the “structural” level of inequality; a level that is due to the intrinsic forces of structural change. Then any deviations -positive or negative- from this level can be identified as “political”, which would provide a clearer picture of the effects of governmental policies on inequality. Overall, the proposed project both contributes to the advancement of European research and addresses one of the Societal Challenges targeted by H2020 programs, which is working towards “Inclusive, innovative and reflective societies”.
In the first work package, following a survey of the literature, a baseline structural change model that incorporates Roy's selection mechanism is developed. Roy’s selection mechanism operates on a simple comparative advantage idea: A worker chooses to be employed in manufactures as long as her wage in this sector exceeds her labor income in services. In other words, she chooses employment on the basis of her comparative advantage. This mechanism enables us to model worker allocation across sectors more realistically than the standard multi-sector models with wage equalization across sectors.

In the second work package, we collected data for the US economy from 1968 to today in order to determine the changes in the overall income distribution, and labor allocations and wage distributions across sectors. We used this data to realistically calibrate the baseline model and see what percentage of the changes in relocation of activity and resources across sectors and the changes in income distribution can be explained by structural change. We show that our model can reproduce 60 percent of the actual change in the employment shares and 38 percent of the worsening in inequality. Although these are significant amounts, the results suggest a big role for factors that are not directly related to structural change, such as demographics, market power, migration, and politics.

Our computations show that the worsening of inequality during structural change is mainly due to the reversal of the roles of the two sectors in high-wage versus low-wage sector status. While the manufacturing sector was the low-wage sector initially, its average wage level became much higher by the end of the structural change period. During the same period, service sector wages move in the opposite direction; average wage level becomes lower as a result of the structural change that brings in many workers that are shed from the manufacturing sector. The bigger size of the now low-wage sector, services, leads to a worsening in the overall income distribution.

In the third work package, we extended the model to make it more realistic in terms of capital income distribution. This meant endogenizing the savings decisions of households in order to account for the pattern of saving rates across the income distribution. Since the literature so far has used the solution to a series of static models to capture the dynamic behaviour of the economies during the structural change period, having a fully dynamic model will advance the state-of-the-art. For this purpose, we set up and solve a three-period OLG model with bequests and Roy’s selection mechanism.

Both the baseline model and the extended version can be simulated not only for US, but also for European countries. As such, the project creates a tool to guide policy makers. One can use it to discuss the trade-offs in creating more inclusive societies during the structural changes any EU-member country is expected to go through.

Dissemination of the project results will be through academic publications and conferences. Aside from the international conferences the researchers will present their results at, there will be a special workshop to disseminate the results of the project and collect feedback from the eminent researchers of the field.
The proposed research will be the first example in literature to analyze structural change in a dynamic general equilibrium model with heterogeneous agents, flexible labor supply and complete asset markets. Structural change models are mainly confined to the representative agent setup. Heterogeneous agent models were restricted to incomplete asset markets until recently. In the last decade, there has been an increase in the number of growth models that use heterogeneous agents in complete asset markets setting. However, these models analyze the income distribution-economic growth relationship at the aggregate level with one-sector growth models and do not deal with structural change. In sum, the infusion of structural change models with heterogeneous agents will be novel and advance both strands of the literature.

One of the Societal Challenges targeted by H2020 programs of the EU is working towards “Inclusive, innovative and reflective societies”. Therefore, the research embodied in this project, which focuses on the relationship between economic development, income inequality and economic policies, is utmost relevant to EU’s strategic priorities. By creating a model that can be simulated for different countries, the project aims to support policy makers in making informed decisions.
US capital income inequality (Gini and Share of Top 5 %), 1968-2007.
Evolution of US Wealth Inequality in PSID vs World Inequality Database.
US employment shares in manufacturing and services, 1968-2007.
US labor income inequality (Gini coefficient), 1968-2007.
Wage distributions across sectors during structural change