Exploring the effects of capital mobility on the saving–investment nexus: evidence from Icelandic historical data

ABSTRACT

We explore the effects of capital mobility on the relationship between saving and investment using historical data for Iceland. First, we analyse the saving–investment (S-I) correlation for the period of restricted capital mobility using data from 1960 and 1994. We then add a period of free capital mobility between 1994 and 2008 and estimate the correlation for the period 1960–2008. Finally, we extend our analysis to the 2008 to 2016 period, when capital controls were imposed in response to the crisis. Institutions matter: We find institutional changes, in particular, Iceland’s entry into the European Single Market in 1994, coincided with a fall in the long-run correlation between saving and investment. However, the correlation weakens further when we include the post-crisis regime of capital controls, suggesting a weaker relationship between savings and investment in this regime. We discuss the possible reasons for this pattern and also the implications of our findings for post-crisis policy in small open economies.

KEYWORDS:

• Feldstein–Horioka puzzle
• capital mobility
• financial crisis

Acknowledgements

The authors are grateful to two anonymous referees for their comments.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 See Benediktsdottir, Danielsson, and Zoega (2011), Hreinsson, Gunnarsson, and Benediktsdóttir (2011), Níelsson and Torfason (2012) and Johnsen (2014) on Iceland’s financial crisis.

2 See, e.g. Oh, Kim, Kim, and Ahn (1999), Katsimi and Zoega (2016), and Raza, Zoega, and Kinsella (2018b).

3 GDP per capita in 2010 dollars was 46,212 dollars in Iceland in 2017, 45,688 in Denmark, 39,504 in Finland and 45,486 in Sweden. Source: OECD.

4 The fishing industry prior to the reduction in subsidies was absorbing more than 40% of government expenditures.

5 The OMX15 covering the 15 largest corporations increased six-fold over the same period and nearly by a factor of nine from its bottom in 2001 to its peak value in 2007. See Aliber (2011) and Halldorsson and Zoega (2010).

6 Source: Registers Iceland (skra.is/Markadurinn/Talnaefni).

7 There is now a very large empirical literature investigating the FH hypothesis. For a survey see, Apergis and Tsoumas (2009) and Kumar and Bhaskara (2011).

8 A variety of explanations have been proposed for the FH puzzle. Coakley, Kulasi, and Smith (1996) argue that large and persistent current account deficits may reduce access to international capital markets. According to Tobin (1983) and Summers (1988), governments may try to avoid deficits for financial stability as well as surpluses.

9 Conversely, non-stationarity of the residuals implies that the variables do not share a common long-run path.

10 See Pesaran, Shin, and Smith (2001) for a detailed discussion.

11 Throughout our empirical analysis, we include a dummy for the potential break points in our models as indicated by ZA test. We only keep the dummy in the model, if found to be statistically significant.

12 Note that the results in Table A4 are only based on simple OLS with no lags. Due to less no. of observations in free capital mobility and capital control regime, it is not possible to estimate a model with lags.

13 This pressure tends to be stronger in countries without sovereign currencies, making them bow to the demand of international creditors as was the case in some Eurozone countries. In most Eurozone countries, the current account balances have rebounded but the economies experienced longer recessions.

14 See Zoega (2010), amongst others.

15 See Raza et al. (2018a) for the comparison of economic recovery in Iceland and Ireland.

Additional information

Funding

This work was supported by the Iceland Centre for Research [grant number 130551-052] and the University of Iceland Research Fund.