New rules, same old paradigm

IN 2013 investigators from America’s Senate shone a harsh light on a highly profitable unit of Apple that was registered in Ireland, controlled from America—and not paying tax in either country. That this “stateless-income” structure was perfectly legal highlighted a big loophole in the global system for taxing multinationals.

There are many such gaps, and the reason is that the patchwork of national rules and bilateral treaties governing how much tax companies owe, and to whom, is horribly dated. It was designed for the manufacturing age. Business today is increasingly digital, services-based and driven by intangible assets, including rights to exploit intellectual property (IP), from patents to logos. These are easier than physical assets to shuffle from subsidiaries in high-tax countries to those in low-tax ones. In short, they make the old rules easier to game.

Hence the relentless rise of tax planning as a core part of multinationals’ business plans. The OECD reckons that the resulting revenue losses to national exchequers have grown to as much as $240 billion a year, or 10% of global corporate income-tax receipts—an estimate it considers very conservative. The share of American firms’ profits that they book in low-tax havens has more than doubled since the 1980s; this has helped reduce the actual tax rates they pay, relative to their home country’s headline rate (see charts). America’s 500 largest firms hold more than $2 trillion in profits offshore. Its tax laws encourage this, because profits its companies make abroad are taxable in America only when repatriated.

Two years ago the Group of Twenty (G20), a forum for big economies, asked the OECD to produce reforms aimed at curbing these corporate-tax gymnastics and ensuring that multinationals were taxed “where economic activities take place and where value is created”. The request was motivated in part by growing public anger over firms not paying their “fair share”, and partly by hunger for more tax revenue in an era of austerity.

The resulting “Base Erosion and Profit Shifting” (BEPS) proposals were released this week. They are the biggest shake-up of multinational taxation since the basics of the current framework were put in place in the 1920s. G20 governments are expected to approve them at a summit in November.

OECD officials were predictably upbeat as they unveiled the plan. Angel Gurría, the organisation’s secretary-general, declared that it will “put an end to double non-taxation”. Pascal Saint-Amans, the OECD’s tax chief, said it marked a “change of paradigm” that should help to make tax planning “marginal” rather than “a core part of business models” (though he accepted there was still much work to do; two years is but a blink of an eye in global tax diplomacy). The reality is less cheering: the project was flawed from the start because it was impossible to achieve consensus in favour of the radical overhaul that was needed. The result is a patch-up job that offers improvements in certain areas but fails to deal with the core problems.

Start with the good bits. Companies will be required to do more country-by-country reporting of where they really earn their revenues, hold their assets and employ people, and where they book their profits—information that is often lacking in their published accounts. This will give tax authorities (though not the public) a clearer picture of how much profit is being shuffled around for tax purposes. National tax authorities will also get more information on “comfort letters” that other countries’ taxmen have provided to companies, blessing their tax arrangements. The European Commission, the EU’s executive arm, is investigating what it suspects are unfair sweetheart deals that the Netherlands, Ireland and Luxembourg have struck with companies ranging from Starbucks to Fiat Chrysler. This week the EU countries got a head start in implementing the OECD’s reform proposals by agreeing on the automatic exchange of information on their cross-border tax rulings.

Double Irish on the rocks

Anticipating new regulations requiring greater transparency, some firms are backing away from tax practices once deemed uncontroversial. Amazon, for instance, has opened taxable branches in European countries where it has lots of customers; no longer are all its profits diverted to low-tax Luxembourg. Some companies are pre-emptively paying more tax: the Luxembourg arm of a large international bank waived a favourable tax ruling in order to raise its effective tax rate, because it feared that paying only 15% might attract negative headlines, admits one of its directors. And low-tax countries are dismantling their most invidious tax-minimisation structures, such as the notorious “Double Irish” (see next story).

However, in some of the most important areas, such as grappling with how to tax cross-border online sales, cans have been kicked down the road. Several proposals were diluted at the insistence of powerful countries (not least America, whose IP-rich multinationals are the main target of the reforms). A case in point is the weak proposals on controlled foreign corporations, subsidiaries used to defer tax by parking it offshore. The OECD’s “action plan” on this issue—one of 15 in its report—is merely a discussion of alternative approaches, none of which is intended to serve as a minimum standard.

In the haggling among governments as the reforms were being drawn up, early progress was reversed in areas such as interest deductibility (a big cause of lost tax revenue, because firms can claim excessive deductions on debts owed to corporate cousins) and the treatment of hybrid instruments (which companies treat as debt in one country and equity in another, to claim two lots of tax deductions). Lee Sheppard, a tax lawyer who followed the talks closely, especially blames American negotiators for weakening proposals or making them unnecessarily complex: “It’s as if BEPS was being killed off from the inside by letting Americans loose on it.”

The biggest disappointment is that, in opting to renovate the existing system, the OECD has stuck with its most deeply flawed pillar: the “independent entity” principle. This rests on the fictitious assumption that the various parent and subsidiary companies in a corporate group act like separate legal persons that transact with each other at arm’s length.

They often do not, because transacting at non-market prices, to shift profits to tax havens, was precisely why some subsidiaries were set up. Such antics are supposed to be kept in check by tax authorities’ rules on “transfer pricing”. But these are complex and often ineffective, in part because it is so hard to be sure if royalties for patents, copyrights and the like have been set at fair prices. The BEPS reforms seek to toughen the rules, but in doing so they add yet more complexity.

Some economists argue that it would be better to do away with corporate income tax altogether, replacing it with higher levies on distributed income, such as dividends. But there are other ways of taxing profits that better reflect economic reality. These involve treating companies as single entities, rather than clusters of supposedly independent parties, and then carving up countries’ taxing rights over the firms’ global profits according to an internationally agreed formula based on sales, assets and other measures.

Under such a system, if it could be agreed on, what would matter most is where the ultimate purchaser of a product is based, something that is harder to manipulate for tax purposes. The approach already works fairly well within some federal countries, for instance in divvying up state taxation of firms that operate across America. “Nobody would ever think of abandoning that system in favour of the international transfer-pricing model, a lesson the OECD would do well to reflect upon,” says Gabriel Zucman, author of “The Hidden Wealth of Nations”, a book about tax avoidance. However, OECD officials say they could not muster enough support for dismantling the old system. America was keen to retain transfer pricing for international commerce.

The next phase could be just as messy. The OECD does not pass laws or sign tax treaties; it merely issues guidelines. Governments are the ones who will have to implement BEPS. Or not. “There is bound to be a significant variation in the timing of implementation and interpretation of how the rules are applied,” says Francesca Lagerberg of Grant Thornton, an accounting firm. Some of the proposals would have to be approved by national parliaments, raising the bar higher still. America’s Congress so far seems unlikely to scrap its vast array of deductions and other benefits, for example.

Some countries, including Britain, with its new “diverted-profits tax”, have pushed through anti-avoidance laws that may be hard to make fit with BEPS principles. That is probably because they have little faith that other countries will implement them in a way that produces a coherent international framework, says Bartjan Zoetmulder of the Dutch tax advisers’ association. If co-ordination is weak, unilateral measures like Britain’s could accelerate as other countries rush to protect their tax bases.

Multinationals’ fear is that growing friction between countries over who gets to tax the lion’s share of their profits leads to the return of double taxation—something that, thanks to the global network of tax treaties, they have been spared from for decades. “There will definitely be more disputes,” says a tax adviser to multinationals. “BEPS is an excuse for all to seek a bigger slice of pie.” The OECD’s reforms include a strengthening of existing procedures that ensure there is no double taxation. But these have yet to be tested. The head of tax at a global software firm fears that country-by-country reporting, as proposed by the OECD, will only encourage countries to dispute their cut of the taxes a given firm is paying.

Drowned out

The developing countries that do not belong to the OECD also view BEPS with mixed feelings. Their gains from curbing tax avoidance could be huge: the IMF estimates that lost tax revenues in poor countries are equivalent to 1.75% of their GDP, three times the level in advanced economies. But their voices were often drowned out in the talks, despite OECD officials’ efforts to ensure they were heard. Some issues that African countries care about, such as the misreporting of customs paperwork, were left out of the talks, says Logan Wort of the African Tax Administration Forum, a group for national tax agencies.

Poor countries may find they lack the institutional capacity to take part in the proposed exchanges between tax authorities of firms’ country-by-country data. So they may still struggle to get data on multinationals operating on their territory—and thus remain ill-equipped to challenge cheeky tax planning.

Unimpressed with the sidelining of developing countries, and with the retention of transfer pricing, campaigners against corporate-tax avoidance have declared BEPS at best a partial success, a first step towards something more substantial. But if, instead of agreeing on how to proceed, governments end up fighting for their cut of multinationals’ profits, a return of double taxation could be in prospect. That might make campaigners happy—some may conclude that multinationals are getting their just deserts for all that tax planning. But it would not be fairer.