An empirical investigation of the financialization convergence hypothesis

ABSTRACT

Claims of global homogenization towards a singular model of finance capitalism constitute a “financialization convergence hypothesis” that has not been subject to systematic empirical scrutiny. Using extensive firm-level data we center on the key indicator of firm leverage, and reveal that substantial cross-national and cross-firm variation still persists. We first compare distributions across OECD countries and find no significant evidence of convergence over time. We then assess whether firms classified as prudent by a simple leverage threshold comprise a declining share of total financial assets over time. We find they do not, and that trajectories remain largely distinct. We do find empirical evidence of financialization convergence in two specific areas. First, there was convergence within the US and the UK in the years immediately proceeding the crisis – but not in other countries representative of stereotypical non-Anglo-American types financial systems, such as Germany and France. Second, we find convergence within the category of large, transnationally-active financial firms. Overall our results suggest that while the behavior of the world's largest globally active financial institutions is converging irrespective of home domicile, their activities are not necessarily leading to the general global homogenization of financial forms and activities implied by the financialization convergence hypothesis.

KEYWORDS:

• Financialization
• global finance
• national systems
• convergence
• financial firms
• banking
• leverage

Acknowledgments

The authors are indebted to Richard Deeg, Iain Hardie, Gregory Jackson, Charlotte Rommerskirchen, Robin Vidra, Sam Knafo, participants at a June 2015 workshop at the University of Edinburgh and a February 2016 workshop at the Freie Universistat, Berlin. We received very helpful comments on early drafts of this paper from all of these people, as well as three anonymous reviewers. This paper was conceived via an International Studies Association Workshop Grant in 2014 and the paper was presented at the 2015 annual meeting of ISA. Remaining errors are our own. The referenced appendix will be available at http://www.wkwinecoff.info/research as well as the journal's repository.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. See also Basel Committee on Banking Supervision (2008, p. 11).

2. A description of the leverage rules in the new global regulatory standard negotiated under the auspices of the Bank for International Settlements’ Basel Committee for Banking Supervision is here. http://www.bis.org/publ/bcbs270.pdf, accessed 15 April 2017.

3. An introduction to the concept of bank leverage, and its importance for bank stability, can be found in Admati and Hellwig (2013, Chapters 2 and 3).

4. Most prevailing economic theory does not provide a clear basis for assuming that economic pressures would produce convergence in behaviors associated with financialization, such as increasing leverage. There are two limitations of dominant economic theory that are important to note. First, many macroeconomic models assume price-taking open economies, where factor prices are equilibrate across countries and the organization of political economies (e.g. according to CME or LME priorities) is not a factor. Second, these economies typically produce and trade without financial markets. Models that include financial actors frequently derive their behaviors within the tradition of the Modigliani–Miller capital structural irrelevance principle, which relies on representative agents and thus assumes little behavioral diversity.

5. For more, see http://www.bvdinfo.com/en-gb/our-products/company-information/international-products/bankscope, accessed 4 January 2016. As extensive as BankScope is, it is not perfect. In the case of duplicate firm observations within a given year, due largely to reporting differences across countries and classes of firms, we included the first reported firm per year for consistency purposes. Key variables such as equity over total assets are standardized reported variables within BankScope. Total real assets were calculated on the basis of reported annual exchange rates and unit levels as reported by BankScope.

6. A further description is found in Appendix 1.

7. It might be tempting to define firms as belonging primarily to market-based or bank-based systems according to type, but financial institutions could engage in market-based bank behaviors regardless of the legal classification. Moreover, regulatory differences across jurisdictions might call into question any ad hoc classification scheme. Thus, we prefer to infer such behaviors from the balance sheet information in our data.

8. We thank an anonymous reviewer for encouraging us to think along these lines.

9. Appendix 3 contains a further graphical description of temporal trends within the countries in our sample.

10. Recent quantitative studies of financialization across countries tend to pool trends across countries, rather than explore them as potentially differentiated systems. For example, Flaherty (2015) also engages in a pooled sample, controlling for country-level variation but ultimately trying to find out the general capital/labour bargaining effects of financialization within the OECD for his 14 country sample. Another more recent manifestation of this kind of analysis is Dünhaupt (2017), which uses a regression model to estimate how the labour share of income changes as a result of financialization. Importantly, the analysis uses country-fixed effects in an attempt to adjust for omitted variable bias given a 13-country sample. We have no quarrel with these studies, which are important advances. Importantly, these are not studies of convergence but rather are seeking to estimate a general effect of financialization on wealth distribution, as a general trend over many countries. Yet the fact that countries can be pooled in such a way suggests the confidence within some existing scholarship that pooling national economic systems is generally uncontroversial.

11. For more, see Clauset et al. (2009).

12. These distributions are not reported for the sake of brevity, but are available from the authors upon request.

13. In the Appendix, we replicate this analysis for other indicators of firms’ balance sheets. The results indicate less distributional similarity.

14. Haldane (2011) also suggested a 5% leverage ratio. We have also looked at higher levels, as American regulatory rules require bank holding corporations to maintain a 6% ratio and systemically important institutions to have 8% equity against assets. The results are very similar to those reported here, so we omit them for the sake of brevity. They are available from the authors upon request.

15. Some of this could be due to riskier firms failing – and thus being driven out of the sample – during the global financial crisis.

16. We also ran KS tests against a distribution of US firms, and these results are reported in the Appendix.

17. We are grateful to an anonymous reviewer for highlighting this point.

Additional information

Notes on contributors

Sylvia Maxfield

Sylvia Maxfield is a professor of Management and Dean at the Providence College School of Business.

W. Kindred Winecoff

W. Kindred Winecoff is an assistant professor of Political Science at Indiana University Bloomington.

Kevin L. Young

Kevin L. Young is an associate professor of Political Science at the University of Massachusetts Amherst.