Dimensions and Determinants of Financialisation: Comparing OECD Countries since 1997

ABSTRACT

The financialisation literature has grown over the past decades. Despite a generally accepted definition, financialisation has been used to describe different phenomena. We distinguish between financialisation of non-financial companies, households and the financial sector and use activity and vulnerability measures. We identify seven financialisation hypotheses in the literature and empirically investigate them in a cross-country analysis for 17 OECD countries and two time periods, 1997–2007 as well as 2008–17. We find different financialisation measures are only weakly correlated, suggesting the existence of distinct financialisation processes. There is strong evidence that financialisation is linked to asset price inflation and correlated with a debt-driven demand regime. Financial deregulation encourages financialisation. There is limited evidence that market-based financial systems are more financialised. Foreign financial inflows do not seem a main driver. We do not find indication that an investment slowdown precedes financialisation. Our findings suggest financialisation should be understood as a variegated process, playing out differently across economic sectors and countries.

KEYWORDS:

• Financialisation
• cross country analysis
• financial deregulation
• property prices

JEL CLASSIFICATION:

• B50
• G20
• P52

Disclosure Statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Ewa Karwowski is Senior Lecturer in Economics at Hertfordshire Business School, and Senior Research Associate at the University of Johannesburg. She worked as an economists for the South African Treasury and as a policy consultant for international organisations such as the ILO and OECD. Her research interests include firm finance, financialisation, and development.

Mimoza Shabani is Senior Lecturer in Economics and Finance at University of East London.

Engelbert Stockhammer is Professor of International Political Economy at King’s College London. He previously held positions at the Vienna University of Economics and Business and at Kingston University. His research interests include post-Keynesian Economics, macroeconomics, financialisation and income distribution.

Notes

1 van der Zwan (2014) provides a similar classification, identifying three main groups of financialisation theories. She, however, distinguishes among financialisation approaches that (1) address changing accumulation regimes, (2) are based on the concept of shareholder value and (3) focus on the financialisation of everyday life. These three categories correspond to our distinction between (1) macroeconomic, (2) mesoeconomic and (3) microeconomic approaches to financialisation.

2 Within the VoC approach Vitols (2014) argues that the effects of financialisation, which he equates with the role of institutional investors, are mitigated by labour market institutions. This means even within VoC financialisation can have different outcomes in different countries.

3 Corbett (1987) argued that borrowing by Japanese non-financial firms effectively had an equity-like character since Japanese banks had considerable control over their clients’ investment decisions. She challenged the dichotomy of bank borrowing and equity issuance by NFCs across countries and questioned whether financial institutions can easily be reduced to bank-based versus market-based systems.

4 Many developing countries, especially in Africa, took on financial liberalisation as part of the IMF’s and World Bank’s structural adjustment programmes.

5 These norms can then of course influence financial activity again like in the case of changing risk perceptions (Besedovsky 2017).

6 Gross operating surplus for the financial sector was considered as alternative measures. Gross operating surplus and value added for the sector are, however, highly correlated. Hence, the measures can be expected to yield very similar results.

7 Financial income data are available for households from Eurostat. Unfortunately, the data only cover 10 of our 17 sample countries and were consequently not included in the analysis.

8 For Canada only one of our five sectoral financialisation measures, household debt, was available. Luxembourg is a small country and an international financial centre. It is therefore not readily comparable to the other OECD countries examined and was excluded as special case from this study. However, it raises an interesting point. Tax havens and international financial centres do feature high in cross country financialisation measures. But the main financialisation theories do not usually treat such cases explicitly.

9 If measures show exactly the same level of financialisation for two or more countries the same rank is assigned to these economies. This is only the case for the financial reforms index and the measure of demand regimes (see Table 3). The financial reforms index is normalised between 0 and 1, which explains why several countries are assigned the value 1, for a completely liberalised financial system. The demand regime indicator is a composite measure composed of two ordinal rankings, which means the same rank can be obtained for more than one country.

10 We have also carried out annual Spearman rank-order correlations to test H1 for the period 1997–2007. The results are essentially the same and available from the authors on request.

11 In our sample of 17 economies four countries have been labelled ‘high’, ‘mhigh’ and ‘low’, respectively, while the groups medium low (labelled ‘mlow’) contains five countries. For the ranking for financial deregulation five countries (Australia, France, Ireland and the UK) have been labelled as ‘high’ because all five have the same average value for the indicator for the period 1997–2007, namely 1.0.

12 These statistically significant correlations remain present when considering annual correlations.

13 The result does not change when considering annual Spearman correlations for the two different periods identified.