ABSTRACT
In view of the Feldstein–Horioka (1980) puzzle, this study examines the relationship between the investment rate and the saving rate using annual data from Norway for 1830 to 2017. The nonlinear version of the autoregressive distributed lag (ARDL) cointegration methodology developed by Shin (2014), based on the linear ARDL bounds cointegration testing approach of Pesaran et al. (2001), is implemented. The findings show that there is a long-term cointegration relationship between the saving rate and investment rate. The nonlinear ARDL model indicates that there are short-run but no long-run asymmetric effects. It is concluded that Norway respects its intertemporal budget constraint and that international capital mobility is high.
Acknowledgments
We thank Patrice Lesgourgues, the editor and anonymous referees for their helpful comments and suggestions on an earlier version. The author remains responsible for any remaining errors.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1 The specification of the alternative Equation (2) with ECM−1 is estimated after imposing the same optimum lag structure as that in Equation (2).
2 Shorter annual series are available in Taylor (Citation1996) or Jordà, Schularick, and Taylor (Citation2017).