Structural Change and the Kaldor Facts in a Growth Model With Relative Price Effects and Non-Gorman Preferences

Abstract

Growth of per-capita income is associated with (i) significant shifts in the sectoral economic structure, (ii) systematic changes in relative prices and (iii) the Kaldor facts. Moreover, (iv) cross-sectional data shows systematic expenditure structure difference between rich and poor households. Ngai and Pissarides (2006) and Acemoglu and Guerrieri (2008) are consistent with observation (i)-(iii) but abstract form non-homotheticities of preferences. However, they cannot replicate the structural change between the U.S. goods and service sector quantitatively. This paper presents a growth model, which reconciles both forces of structural change -relative price and income effectswith balanced growth on the aggregate. The theory is simple and parsimonious and contains an analytical solution. The model can replicate shape and magnitude of the nonbalanced sectoral facts as well as the balanced nature of growth on the aggregate. In a structural estimation, the model's functional form is exploited to obtain estimates for the relative importance of income and price effects as determinants of the structural change.Keywords: Structural change, relative price effect, non-Gorman preferences, Kaldor facts. JEL classification: O14, O30, O41, D90. * I would like to thank Josef Falkinger, Reto Föllmi, Volker Grossmann, Marcus Hagedorn, Danyang Xie, Fabrizio Zilibotti and Josef Zweimüller for illuminating discussions. Moreover, I would like to thank Gregori Baetschmann, Victoria Galsband, Sandra Hanslin, Andreas Müller, Nick Netzer, Iryna Stewen, Raphael Studer and Franziska Weiss for many valuable comments and suggestions.

Dynamic Status Effects, Savings, and Income Inequality

“…To address the reality that the saving rate increases with wealth over time, the literature mainly considers technological factors that increase the return to capital over time and, in turn, the saving rate. But by associating the increased saving rate with increasing returns to capital, this explanation contradicts recent evidence provided by Boppart () and Ledesma and Moro (), suggesting that the return to capital is decreasing over time.…”

The Drivers of Structural Change

“…Focusing on the postwar period by means of USA data, Boppart () – who incidentally was the very first to present a growth model with intertemporal optimization that reconciles the Kaldor facts on aggregate dynamics with structural change driven by income and relative price effects – exploits the functional form of his multi‐sector model of growth and finds that the income and relative price effects have roughly equally influence structural change towards services. Confirming the significant contribution of both mechanisms to structural change, Comin et al .…”

“…A first facet of the literature on structural change shows that the process of reallocation of economic activity across sectors can be due to changes in the structure of demand resulting from changes in real income (e.g. Pasinetti, 1981;Falkinger, 1994;Echevarria, 1997;Laitner, 2000;Zweimüller, 2000;Caselli and Coleman, 2001;Kongsamut et al, 2001;Gollin et al, 2002Gollin et al, , 2007Greenwood and Seshadri, 2002;Meckl, 2002;Steger, 2006;Bonatti and Felice, 2008;Foellmi and Zweimüller, 2008;Duarte and Restuccia, 2010;Boppart, 2014). Structural change, therefore, is driven by non-homothetic tastes, such as the Stone-Geary preferences, which generate non-linear Engel curves.…”

“…Meijdam et al ., ; Stijepic, ), the income distribution (e.g. Hein and Vogel, , ; Hartwig, , ; Boppart, ), the level of public intervention (e.g. Stijepic and Wagner, ; Felice, ) and the total amount of time devoted to leisure (e.g.…”

“…As this condition is empirically verified for different sets of consumption sectors in the literature, see for example Ngai and Pissarides (), Buera and Kaboski (), Herrendorf et al . (), Boppart (), Comin et al . () and Swiecki () – Acemoglu and Guerrieri () considers the case in which the elasticity of substitution is relatively low as the empirically relevant case.…”