Does Centralisation of FX Derivative Usage Impact Firm Value?

• Date: 01 March 2015
• Published date: 01 March 2015
• Author: Håkan Jankensgård
• DOI: http://doi.org/10.1111/j.1468-036X.2013.12014.x

Does Centralisation of FX Derivative

Usage Impact Firm Value?

Håkan Jankensgård

Department of Business Administration, Lund University, P.O. Box 7080, 220 07, Lund, Sweden

E-mail: hakan.jankensgard@fek.lu.se

Abstract

Previous research has shown that firms identified as derivative users tend to be

valued at a premium relative to non‐users. In this paper I develop the hypothesis

that the ‘derivative premium’is higher in firms with centralised FX exposure

management, compared to a decentralised approach in which subsidiaries retain

bank contacts and/or decision‐making authority. This study benefits from unique

survey data on the FX management practices and derivative usage of Swedish listed

firms. The data supports the centralisation‐hypothesis. Firms with a centralised

approach have a statistically significant derivative premium of around 15%,

whereas there is no premium for decentralised firms.

Keywords: centralisation, risk management, currency risk, derivative, hedging

JEL classification: G30, G32

‘…many of the Company’s subsidiaries are exposed to currency risk…To manage such

currency risk, the Company’s policy requires its subsidiaries to hedge their foreign currency

exposure from binding sales…’

ABB, annual report 2009

‘Decisions regarding the hedging of operating cash flow exposures are made by the

individual Business Areas in collaboration with Group Treasury.’

Trelleborg, annual report 2009

The author gratefully acknowledges several valuable comments and suggestions by an

anonymous referee. I furthermore wish to thank John Doukas (the editor), Raj Aggarwal,

Tom Aabo, Anders Vilhelmsson, Lars Oxelheim, Jens Forssbæck, Fredrik N.G. Andersson,

and Anders Löflund for helpful comments. I thank the Jan Wallander and Tom Hedelius

foundation and the Tore Browaldh foundation for financial support. Any remaining errors

are the responsibility of the author alone.

European Financial Management, Vol. 21, No. 2, 2015, 309–332

doi: 10.1111/j.1468-036X.2013.12014.x

© 2013 John Wiley & Sons Ltd

‘Currency risks in the subsidiaries’net exposures are hedged according to the policy of each

individual subsidiary.’

Ratos, annual report 2009

1. Introduction

Conventional wisdom holds that there are significant benefits from centralising foreign

exchange (FX) exposure management. By centralising a firm can avoid duplicating costly

risk management activities, achieve lower transaction costs, and concentrate its financial

competence. Yet empirical surveys (Oxelheim, 1984; PricewaterhouseCoopers, 1995;

Edelshain, 1997; Bodnar et al., 1998, 2003; Belk, 2002) suggest that in practice

decentralised exposure management, i.e. FX derivative usage by business units other than

central Treasury, occurs on a non‐trivial scale.

This paper is the first to analyse if the so‐called ‘derivative premium’is conditional on a

centralised approach to FX exposure management, as opposed to a decentralised approach

in which business units retain bank contacts and/or decision‐making authority. The

derivative premium refers to the finding that firms identified as derivative users tend to be

valued at a premium relative to non‐users, typically in the range of 3–16% (Allayannis and

Weston, 2001; Carter et al., 2006; Kim et al., 2006; Clark and Judge, 2009; Allayannis

et al., 2011).

1

Using unique survey‐data on the FX derivative practices of Swedish listed firms, the

estimated premium associated with derivative usage for firms with a centralised approach

is around 15% and statistically significant at the 10%‐level. Firms with decentralised FX

exposure management, on the other hand, are not associated with a derivative premium.

This suggests that FX derivative usage is valuable to a firm only if it is centralised. We

thus have evidence of a ‘centralisation premium’, which is robust across different model

specifications and different definitions of decentralisation.

These findings are consistent with the received wisdom that decentralised exposure

management is associated with a number of coordination problems and cost inefficiencies

(the efficiency explanation). Another possibility is that a decentralised approach suffers

from higher agency cost of risk management, i.e. an excessive use of derivatives that cater

primarily to the demand of risk‐averse managers (the agency explanation). A third

possible explanation is that firms with decentralised FX exposure management have more

decentralised decision‐making generally, and that derivative usage is less valuable given

such a business model (the business model explanation). In particular, subsidiaries in

decentralised companies are more likely to have relatively independent capital structures

and bankruptcy risk. Derivative usage could fail to add value in such firms because

1

In this literature this finding is commonly referred to as the ‘hedging premium’. When, as in

this paper, the data does not allow us to distinguish between the hedging and speculative

motives for derivative usage it is more appropriate to use the term ‘derivative premium’. Not

all papers have found a derivative premium. Jin and Jorion (2006), for example, find no such

evidence for a sample of US oil and gas firms.

© 2013 John Wiley & Sons Ltd

310 Håkan Jankensgård