The dynamics between the stock market and exchange rates: Spain 1999–2015

Abstract

Despite the significance of the subprime crisis, there are few studies of its impact on the dynamics between stock markets and exchange rates in Eurozone countries. This study helps to remedy that shortage by analysing the dynamics between the stock market and exchange rates for the Spanish economy in the period 1999–2015 with sub-periods 1999–2007 and 2008–2015, both before and after the financial crisis. We analyse the Granger causality between the Spanish stock market and real effective exchange rates and EUR/USD, EUR/JPY, EUR/CNY and EUR/GBP bilateral rates, through the Toda and Yamamoto procedure. To check robustness and sign in the direction of causality we use impulse-response analysis. On the one hand, the results show that the relationships analysed are significant only in the crisis sub-period (2008–2015), in which bilateral exchange rates lead fluctuations in the stock market while the latter leads the real effective exchange rate. On the other hand, for bilateral exchange rates the directions that show the impulse-response analysis are consistent with those shown in the Granger-causality analysis and the sign coincides with the data on the merchandise trade balance with the countries in question.

KEYWORDS:

• Stock market
• exchange rates
• Granger causality
• financial crisis

JEL Classifications:

• F31
• F41
• G15

Acknowledgments

We are grateful to David Camino and UC3M for their support in the research process of data collection, FESIDE foundation and participants at the XXXII AEDEM annual meeting and Business Institute (UPV-EHU) workshop for their comments and suggestions. We acknowledge particularly those made by two referees and the associate editor, which contributed to the overall improvement of the paper. Finally, the authors thank Chris Pellow for his English support.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 A currency common to a group of countries is equivalent to fixed exchange rates.

2 Fleming (1962) and Mundell (1963) are seminal papers in the ‘traditional approach’. However, Dornbusch and Fisher’s article is the most cited in relation to this approach. However, this paper and others like Frenkel (1976) and Bilson (1978) must be better framed in the ‘asset view’. They are integrated into the ‘flow approach’ because they share the conclusions regarding the direction of the causality with those of the ‘traditional flow approach’ that dates back to the decade of the 1960s.

3 The Bank of Spain has not published statistics by country for the current account of the balance of payments until 2013. Moreover, these statistics do not show a breakdown by country except in some cases and are not entirely accurate. However, there are very robust statistics of the merchandise trade balance by country. We understand that it is appropriate to use the latter because the trade of most of the companies in the Spanish stock market (above 70%) is based on goods.

4 For EUR/CNY the data start from February 2000.

5 As a sensitivity exercise, to detect non-linear Granger causality between the variables considered, we applied the NlinTS package in R, which performs a non-linear test of causality using feed-forward artificial neural networks. The results are fairly similar, as non-linear causality relationships in general are also stronger in the crisis sub-period. The results of this analysis are available from the authors upon request.

6 In the case of the UK there is information on the current account balance since 2013 and it is an export balance year after year as in the case of the merchandise trade balance, which supports the use of the merchandise trade balance as a ‘proxy' of the current account balance.

Additional information

Funding

This work was supported by Basque Government under grant IT1336-19.