Housing is NOT ONLY the Business Cycle: A Luxemburg-Kalecki External Market Empirical Investigation for the United States

ABSTRACT

We study the residential investment-economic activity nexus in the United States during the period 1960–2020. We find evidence of symmetric and asymmetric frequency-domain Granger causality running unidirectionally from residential investment (RES) to output. This unidirectional causal relationship is both permanent and transitory: transitory shocks in RES have transitory effects on GDP, while permanent shocks in RES have permanent effects on GDP. Our results validate the hypothesis of Fiebiger [2018. ‘Semi-Autonomous Household Expenditures as the Causa Causans of Postwar US Business Cycles: The Stability and Instability of Luxemburg-Type External Markets.’ Cambridge Journal of Economics 42 (1): 155–175] and Fiebiger and Lavoie [2019. ‘Trend and Business Cycles with External Markets: Non-Capacity Generating Semi-Autonomous Expenditures and Effective Demand.’ Metroeconomica 70 (2): 247–262], who state that housing investment in the US can be analogous to a Luxemburg-Kalecki external market. Our findings can also be read through the lenses of the recent autonomous demand-led growth literature. In particular, we single out a specific component of autonomous demand and describe its prominent role in the US variety of capitalism. Thus, we conclude that residential investment, despite constituting a small overall share of GDP, is not only the cycle but is also the trend of the US economy.

KEYWORDS:

• Effective demand
• growth
• residential investment
• autonomous demand
• US economy

JEL CLASSIFICATION:

• B51
• E11
• E12
• O41
• O47
• O51

Acknowledgements

We thank two anonymous referees, Brett Fiebiger and the participants in the seminar of the Doctoral Program in Economic History (UB-UC3M-UV) on May 14, for useful and stimulating comments, critiques and suggestions. It goes without saying that the usual disclaimer applies.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 We are referring, here, to the Sraffian supermultiplier model developed by Serrano (1995), Bortis (1997), and Dejuan (2005), and some recent versions of the neo-Kaleckian model (see for example Allain 2015, 2019; Lavoie 2016; Fiebiger and Lavoie 2019).

2 On the other hand, Arestis and Gonzalez-Martinez (2015b) study the relationship between housing and the labour markets, focusing on the role of labor precariousness.

3 The interested reader can refer to Freitas and Serrano (2015) and Lavoie (2016) for in-depth treatments of the issue. The autonomous demand-led growth literature is in line with several non-mainstream economics strands, such as Duesenberry’s (1949) relative income hypothesis, the endogenous money approach and the Luxemburg-Kalecki external markets literature (Fiebiger and Lavoie 2019, p. 247).

4 The ‘Sraffian’ label points to the fact that income distribution is treated as an exogenous variable, rejecting therefore any mechanical link between distribution and economic growth.

5 See Palley (2019) and Fazzari, Ferri, and Variato (2020), which introduce into this framework an explicit consideration of labor market dynamics. See also Hein (2018) for a model depicting an economy whose growth is driven by the evolution of autonomous government expenditures, but without convergence to normal utilization.

6 Fiebiger and Lavoie (2019) and Petrini and Teixeira (2020) explicitly explain how the SSM adjustment mechanism can give insight into how cycles and secular trends are determined by the dynamics of dwellings.

7 For a deeper discussion of endogenous money, the reader can refer to, among others, Palley (1997), Fontana and Setterfield (2009), Cesaratto (2016), and Deleidi and Fontana (2019).

8 See also Duesenberry (1958) on this. The American economist also emphasizes the role played by speculative investment in disconnecting the dynamics of residential investment from that of production income. This provides a possible theoretical explanation for the non-linear nature of the relationship between the two variables, a feature confirmed by our empirical findings in section 5.

9 A supermultiplier mechanism is capable of giving insight into the effects of secular trends in dwelling investment on the secular growth rate (Fiebiger and Lavoie 2019, p. 248).

10 The first article of the list deals with exports, the three following papers focus on government innovation expenditures, while the last one considers a growth model led by debt-financed consumption of workers.

11 We do not include in this list consumer credit-financed expenditures. In doing this, we follow Girardi and Pariboni (2016), who find consumer credit to be influenced by the economic cycle and hence not suitable for consideration as an autonomous component of demand.

12 We are thankful to Jörg Breitung for this suggestion.

Additional information

Funding

This work was supported by the research grant FPI/1866/2016, financed by the European Social Fund, and by the project PGC2018 093896-B-100, granted by the Spanish Ministry of Economy, Industry, and Competitiveness (MINECO), which is gratefully acknowledged.