Outlook: Tax loopholes

Click to follow
The Independent Online
IT SAYS a lot about the complexity of stamp duty reserve tax that mere whisper of it sends even hardened specialists scurrying for cover. Yet, arcane as it may be, it is a nice little earner for the Exchequer. Projected revenues of pounds 2.4bn for the 1998-99 financial year were enough to prompt an Inland Revenue swoop at the weekend to plug a loophole that was in danger of looking like a yawning chasm.

What was a dark corner of the tax world was suddenly bathed in light, when BP and Amoco escaped paying $800m in duty during their merger last year by issuing new shares as foreign currency bearer shares instead of American Depositary Receipts, through which US investors usually deal in UK equities. The saving arose because the bearer shares were made exempt from the 1.5 per cent stamp duty reserve tax in 1989 in order to assist companies wanting to borrow overseas.

The Revenue says that, while this purpose will continue to be allowed, it has cracked down on transactions which the legislation was not designed to cover. However, though the ruling applies from midnight last Friday, it is apparent that it has failed to stop similar tax savings by two other high-profile cross-border deals involving UK companies - the $35bn merger of pharmaceutical company Zeneca and its Swedish rival, Astra, and Vodafone's $62bn purchase of AirTouch.

In taking this swift action, Patricia Hewitt, Economic Secretary to the Treasury, is probably on to a winner. Though the tax hauls are large, the sums are unlikely to be enough to - by themselves - to stem the tide of cross border mergers.

Hence, the Chancellor can look forward to a steady and easily collected revenue flow - which is presumably why the Conservatives never got around to acting on the plan announced in 1990 to abolish stamp duty on shares and securities all together. Some globally-minded chief executives may complain it puts the UK at a disadvantage. But if a merger can only be made to work because of its tax efficiencies, then it is almost certainly not worth doing in the first place.