Growth and Cycles of the Italian Economy Since 186

Authors: Fabio Clementi Marco Gallegati Mauro Gallegati

Publish Date: 2014/12/05

Volume: 1, Issue: 1, Pages: 25-59

PDF Link

Abstract

Based on a newlyavailable large set of historical national accounts the paper revisits the main features of economic growth and cycles in Italy for the postUnification period 1861–2011 Alongside the structural changes in growth dynamics the main sources of output and productivity growth are identified As regards the analysis of the underlying cyclical component a business cycle chronology is first established and then both the specific patterns of individual cycles and the comovements of output with key macroeconomic variables are investigated In the 150 years since its political Unification Italy’s economic growth was mainly propelled by consumption and investments whereas on the supply side the industry and services sectors were by far the main contributors also because of the positive effect of labour reallocation to nonfarm activities Over the same period Italy experienced approximately 20 business cycles of varying duration and amplitude Output fluctuations were dominated by the shortterm variability of agricultural production before World War II and by fluctuations of the industry sector thereafter The cyclical behaviour exhibited by aggregate demand components conforms quite well to that evidenced in the standard international business cycle literature although some exceptions arise in the preWorld War II yearsEarlier drafts of this paper were presented at the study day in honour of Guido M Rey “L’economia italiana modelli misurazione e nodi strutturali” Università degli Studi Roma Tre 5 March 2010 and the conference “Sviluppo economico e benessere” Università Politecnica delle Marche Ancona 5 and 6 November 2010 We are grateful to all the participants for their useful comments We also thank Alberto Baffigi for his precious help with the historical sources used and Claire Giordano and Francesco Zollino for sharing their data with us upon request Last but not least we thank two anonymous reviewers for their thorough review and highly valuable comments and suggestions which significantly contributed to improving the quality of the paper The usual caveats applyItaly has a solid and long tradition of studies dealing with the issue of reconstructing national accounts in the postUnification period The first set of historical national accounts from 1861 to 1956 due to the National Institute for Statistics ISTAT 1957 was revised and improved by a group of researchers of the University of Ancona under the supervision of Fuà 1969 Maddison 1991 proposed a revision of the GDP series for the period 1861–1989 by lowering the initial levels of the ISTATVitali GDP series and then Rossi et al 1993 reconstructed the GDP series from the expenditure side for the period 1890–1990 using the new benchmark for 1911 and the new estimates provided respectively by the Bank of Italy Golinelli and Monterastelli 1990 Rey 1991 Recently Fenoaltea 2005a b 2006 has produced for the period between national Unification and World War I the first entirely new estimates of Italian aggregate GDP since ISTATVitali by combining Federico 2003 series for agriculture with his own series for industries and services Finally Baffigi 2013 has provided new GDP series together with supply and demand side estimates covering the 150 years after Italy’s political Unification as a part of the “150 anni” national accounts project that includes the Bank of Italy ISTAT and the University of Rome “Tor Vergata”—as well as academics from other institutions1The availability of historical macroeconomic time series has favoured a proliferation of studies especially in recent years on the nature and causes of business cycle fluctuations2 and—in a historical perspective—on growth in Italy3 As a result in the last two decades the knowledge about the historical patterns of Italian economy over such a long time span has considerably improved From this point of view the availability of new data provides the researcher with the opportunity to look at old facts and interpretations in new ways and to answer the question to what extent these new estimates provide contrasting results about growth and cycles with respect to those presented in the previous literature especially in the preWorld War II periodIn this paper we revisit the Italian economic growth and cycles in the postUnification period using the new historical accounts presented in Baffigi 2013 for the period 1861–2011 In particular we analyze whether the revised estimates provide new evidence as to the presence and number of structural changes in GDP growth that can suggest a different alternative interpretation of the phases of Italy’s longrun economic growth especially with regard to the interpretation that traces back Italy’s economic development to the beginning of the nineteenth century Specifically following the recent literature supporting the view that macroeconomic time series and GDP in particular can be represented by stationary fluctuations around a deterministic segmented trend eg Rappoport and Reichlin 1989 we test for multiple structural breaks at unknown dates in the trend function of GDP growth rate as suggested by Bai and Perron 1998 2003 Moreover since the new dataset also provides us with new estimates for both the supply and demand sides this paper presents an analysis of the impact of these revisions to the overall pattern of the various supply and demand components’ contributions to real growthAs regards the analysis of the underlying cyclical component we take into account both the classical and modern definitions of business cycle where the first analyzes the specific patterns of individual cycles and the latter the comovements between GDP and different key macroeconomic variables Although most macroeconomists share the real business cycle RBC view that the “business cycles are all alike” Lucas 1977 we try to go beyond this approach by integrating modern RBC analysis with a detailed description of individual phases and cycles according to the National Bureau of Economic Research NBER traditional approach4 The aim of our methodology is to provide empirical evidence on regularities and discontinuities in economic fluctuations and to reconsider the hypothesis of business cycles similarity of the predominant view In particular following the modern definition of business cycle5 after isolating the underlying cyclical component corresponding to fluctuations of approximate length between 15 and 8 years we analyze the empirical regularities observed in the comovements among different aggregative time series Moreover following the Burns and Mitchell 1946 classical definition of business cycle analysis which suggests to look at the characteristics of each individual cycle and phase in terms of duration amplitude and steepness we also check whether the revisions alter the key features of business cycle fluctuations including the dating of turning pointsThe paper is organized as follows Section 2 describes the new dataset and provides a brief comparison with previous estimates The long run dynamics of Italian economic growth is analyzed in Sect 3 by identifying structural breaks in GDP growth rate along with the contribution of the various supply and demand components to economic growth Section 4 presents the results of the analysis of the cyclical component of Italian GDP using both the classical and modern definitions of business cycle Finally Sect 5 concludes

Keywords: