Tobin tax and trading volume tightening: a reassessment

Abstract

This article extends the previous literature on the Tobin tax. We find that very roughly, a doubling in transaction costs would reduce trading volume by 25% to 40% in the Forex. Most importantly, this article is the first contribution to specify the trading volume of the Forex through different (low and high volatility) regimes. Our results show evidence of nonlinear patterns for trading volumes and transaction costs on the Forex. Thus, the Tobin tax would not have a monotonic impact on trading activity across market conditions. The change in elasticity between low and high volatility regimes would be slight but significantly different. We may suggest that the high-variance regime might be the fundamentalist regime and the low-variance regime might be the chartist regime. It is a first step towards understanding which categories of agents would react to the introduction of a tax. Our results seem consistent with Tobin’s underlying thinking; since a tax would penalize chartists more than fundamentalists, it could reduce exchange rate volatility.

Keywords:

• Tobin tax
• trading volume
• Forex
• transaction costs
• global financial crisis
• Markov switching

JEL Classification:

• E44
• F31
• C22
• C13
• C52

Notes

¹ There are however two empirical studies of the effect of a Tobin tax on volatility: Aliber et al. (2003) and Lanne and Vesala (2010).

² See Hartmann (1999) for an empirical paper about the determinants of Forex spreads.

³ For instance, Hartmann (1999) chose ARIMA (9,1,1).

⁴ We choose 60 business days in line with Galati (2000).

⁵ We also estimate a Smooth Threshold Regression (STR) model to check the robustness of our Markov results. Considering the STR estimates (results are available upon request from the authors), traders were less or irrationally sensitive to transaction costs from September 2008 to April 2009 but are very sensitive (strong negative elasticity to transactions costs) in the other periods.