New Paper Makes Powerful Case Against Tax Treaties

By Nicholas Shaxson, the author of Treasure Islands, an award-winning book about tax havens. Originally published at Tax Justice Network

Tax treaties are an arcane but important part of the international trade and investment system. When a business from one jurisdiction invests in another, the question then arises as to which jurisdiction gets to tax which bits of the income that the investment generates. So countries have for years signed Double Tax Treaties or Double Tax Agreements (DTAs) with each other, to sort out these and other questions. Since the global treaty system began to emerge (after Austria-Hungary signed one with Prussia in 1899), the core aim of the system’s designers has been to make sure that multinationals don’t get taxed twice on the same income: so-called ‘double taxation’. Countries sign them because they think they will attract and smooth the flow of inward investment. It will make their country more ‘competitive,’ the thinking goes.

All of which sound like perfectly reasonable ideas. But of course, beneath these reasonable ideas there’s a world of possible mischief.

The first question is: who designs these tax treaties? Well, the standard answer is this: the OECD. Rich countries. They design the model treaties, which countries then incorporate into their national legislation. In treaty-speak there are “residence countries” (where multinationals are based or ‘resident’, nearly always rich countries) and “source countries”, where the income is sourced: that is the recipient of the inward investment, which is often a poorer country. No prizes for guessing which principle — source or residence — is dominant, in terms of taxing rights. Yes, the rich countries have rigged the game against the poorer countries, and the latter grouping end up with ugly terms that stop them being able to properly tax multinationals that invest in their countries. And, as we’ll see, these treaties are at the end of the day pretty much useless at attracting inward investment.

It’s important to understand that the rich countries haven’t exactly rigged the game to benefit themselves, though: they have rigged them in favour of multinationals. They have turned the quest to eliminate ‘double taxation’ into the pursuit of double non- taxation: often by letting them structure their transactions via mucky tax havens like Ireland, Luxembourg, Switzerland, the Netherlands, or (for the more adventurous) Mauritius or Singapore — all of which specialise in signing reams and reams of these treaties, so that they can serve as conduits and cream off fees for providing tax-cutting (and sometimes secrecy-creating) services.

There is, it is true, an alternative forum to the OECD: the far more geographically representative UN Tax Committee – but they are (due to OECD countries’ power at the UN) woefully under-resourced; under constant pressure to kowtow to the OECD’s models (which they generally do), and roundly ignored. (Actionaid has done a good policy brief on this; see more here.)

Of course the theoretical potential number of tax treaties is not much less than the total number of countries, squared (that is, if each country signed with each other country) – so there are literally thousands of these pesky things out there. Just think of all the consulting fees that the Big Four accountants are able to extract from this racket.

Now in 2013 the Tax Justice Network published an article entitled Lee Sheppard: Don’t sign OECD model tax treaties! which looked at a presentation by one of the U.S.’ top experts in international tax. Her fiery presentation contained gems such as:

“The treaties protect multinationals primarily. That’s all they were ever for: to make life comfortable for multinationals” . . . The international consensus is “basically a load of nonsense that protects multinationals. . . The OECD primarily protects the interests of the United States and the United Kingdom. Even Germany doesn’t get a look in.”


“When you sign an OECD model treaty, you say there is no withholding, or hardly any withholding, on outflows of cash to multinationals. Now why in hell do you want to sign that?”

And here’s another thing. You really, really shouldn’t be signing an OECD double tax agreement with a tax haven. When you do that, you not only prevent your own country from taxing any multinationals effectively – but you also then find that the tax haven is used to create a tax-free pathway (for your locally-sourced multinational income) through the international tax system, escaping tax on that income altogether (see here for a slightly fuller explanation.). Signing a DTA with a tax haven is a recipe for multinationals to ransack your threadbare larder. Indian officials know very full well that India made a ghastly mistake signing a treaty with Mauritius, a secretive, aggressive, corrupt and very fast-growing tax haven in the Indian Ocean. But such is the power of vested interests that it’s rather hard to get rid of these things, once established.

Some developing countries do seem to be breaking away from the consensus, at least. But many poorer countries are still having the wool pulled over their eyes. Why, just between May 2014 and May 2015 Mauritius either negotiated, signed or ratified tax treaties with Swaziland, Malta, Egypt, Kazakhstan, Congo, Rwanda, Algeria, Canada, Czech republic, Hong Kong, Iran, Lesotho, Malawi, Montenegro, Portugal, Tanzania, Saudi Arabia, Vietnam, Yemen, Guernsey, Kenya, and Ghana. What on earth are sleaze-addled Kazakhstan or impoverished Malawi doing signing away taxing rights (and creating additional opportunities for secrecy-oiled shenanigans) by entering into a DTA with a mucky tax haven like Mauritius? It certainly won’t be stimulating a bunch of good new investment, as the section below shows.

Well, now a new(ish) paper by tax scholars Kim Brooks (Dalhousie University) and Richard Krever (Monash University) makes the case against tax treaties in rather more academic, but equally powerful terms. This review by Jan Van de Poel, below, was first published by the Tax Justice Network.

Guest blog: tax treaties may be “a true poisoned chalice for developing countries

In a globalised economy, more than 3000 bilateral tax treaties determine how and how much tax developing countries are able to collect from profits that are shifted across borders by multinationals. Whether these treaties support developing countries’ capacity to finance their development needs is subject to intense debate in the development community. In a July 2015 contribution to IMF’s Victor Thuronyi and Geerten Michielse Series on International Taxation (published by Wolters Kluwer), tax scholars Kim Brooks and Richard Krever provide convincing arguments that such treaties ‘may be a true poisoned chalice for developing countries’.

Brooks and Krever’s paper reads as a comprehensive ‘myth buster’ on tax treaties and cannot be underestimated as a resource for activists and other experts on the matter.

The orthodox rationale for tax treaties is to prevent “double taxation” of multinationals by reallocating taxing rights between an investor’s home jurisdiction and the host jurisdiction, and to allow for effective cooperation between countries in tax matters. Any foregone tax revenue resulting from the ‘reallocation’ of taxing rights is said to be offset by other benefits such as additional inward investment or ‘strategic benefits’ such as access to development aid or security services.

Brooks and Krever’s paper tears these pro-tax treaty arguments apart, making the case these benefits can be better achieved by unilateral action by the source state without giving up taxing rights and losing much needed revenue. Their paper reads as a comprehensive ‘myth buster’ on tax treaties and cannot be underestimated as a resource for activists and other experts on the matter. As it is incredibly rich with arguments and ideas, referring to the most recent literature and political events in the field, I will limit myself to just a few that I found particularly refreshing, as they do not come up often in discussion.

Brooks and Krever tackle the persistent myth that tax treaties are an essential driver for inward foreign direct investment in developing countries. Previous econometric analysis has shown only a very weak direct relation between treaties and FDI, and suggest most of the additional investment is ‘virtual’ as it reflects the routing of income through tax havens as part of multinationals’ tax planning strategies. Brooks and Krever even go one step further arguing that reducing withholding taxes on business income earned in a host jurisdiction may affect investment negatively.

Given that a witholding tax is not a tax on current profits and can be deferred indefinitely by firms willing to reinvest in the jurisdiction, higher withholding tax rates might actually encourage investment: a fascinating point.

Another myth that both scholars address is developing countries cannot do without tax treaties. After developing countries have made their cost-benefit analysis and have decided whether it is appropriate to sign a treaty or not, a second question needs to be answered. Are these benefits possible without signing a treaty? Ensuring exchange of information or dispute resolution through arbitration requires state-to-state cooperation of some sort, but not necessarily a full-blown tax treaty. Increased investment and trade or non-discrimination between foreign and domestic investors is probably best done by building effective tax administrations and domestic legal measures. So, in most instances, the answer is yes: most benefits can be achieved by unilateral measures and do not require the revenue losses associated with signing a tax treaty.

For those who have developed a (dis)taste for all things BEPS (“Base Erosion and Profit Shifting,” the OECD’s project to tackle international tax avoidance) over the past two years, the paper provides additional arguments for not buying into the story that BEPS has fixed the international tax system. Tax treaties make it nearly impossible for source countries to tax business profits derived by a non-resident with no enduring presence or high profile in the jurisdiction.

[An aside: Sheppard explained it like this: “When you sign an OECD model treaty you also sign onto a concept called Permanent Establishment. “It is a rather nonsensical concept that says, ‘well, if you, multinational, are operating in a country and making money in a country, but you have any presence that is short of, oh, a full automobile assembly plant, then you are not taxable in that country at the level of the owner of this plant. This thing has a little circle drawn around it, and it cannot be taxed in a normal way. . . . you do not want to sign a document that has got that in it.

The BEPS proposals attempts to revisit the concept of ‘permanent establishment’ but only marginally addresses this problem. Again, Brooks and Krever, conclude this issue is ‘best adopted through unilateral measures that set boundaries based on source country’s actual tax administration capacity’.

A must-read for everyone interested in the relationship between tax treaties and development, download the full paper here.

The Troubling Role of Tax Treaties, by Kim Brooks (Dalhousie Dalhousie University – Schulich School of Law; Monash University – Faculty of Law; and Richard Krever, Monash University – Department of Business Law & Taxation, July 1, 2015. In Geerten M. M. Michielse & Victor Thuronyi, (eds.), Tax Design Issues Worldwide, Series on International Taxation, Volume 51 (Alphen aan den Rijn: Kluwer Law International, 2015), 159-178.


The notional purpose of tax treaties is to prevent double taxation and tax evasion. The actual purpose is to reallocate taxing rights between an investor’s home jurisdiction (the residence state) and the host jurisdiction (the source state). The effect is to reduce or remove the taxing rights of a source state (a capital importing state) to leave more room for tax in the residence state (a capital exporting state). The revenue costs of agreeing to reduce taxing rights in a treaty are thought to be offset by other benefits. The benefits may be exaggerated. To the extent they may actually be realized, all can likely be achieved more efficiently through unilateral action by the source state.

An excerpt from the paper itself:

Brooks Klever


Endnote: the researchers note that the literature doesn’t generally foster worthwhile investment. There’s another big factor which is particularly important: a lot of the ‘investment’ that takes place is in the form of ’round-tripped’ where local investors take their wealth offshore, dress it in offshore secrecy and then return it disguised as foreign investment, in order to obtain treaty benefits and other goodies only available to ‘foreign’ investors. Much if not most of the literature doesn’t take account of this stuff – so the picture is even worse than it might appear on the surface.

Print Friendly
Tweet about this on Twitter0Digg thisShare on Reddit0Share on StumbleUpon0Share on Facebook0Share on LinkedIn0Share on Google+0Buffer this pageEmail this to someone


  1. Synoia

    Let me leaven this topic with an example:

    In the 1970s I worked for IBM in South Africa. A Mainframe computer was more expensive when sold by IBM South Africa than when sold by IBM Domestic (US).

    The South African staff was paid about 50% of the US’ wages, The Mainframes all came form two Manufacturing plants, Poughkeepsie for the US and one in Germany for Europe, The Middle East and Africa..

    The price difference was a constant struggle with customers.

  2. EForster

    The money to pay those employed, and to settle their payroll taxes, along a supply chain, is derived ultimately from the total sales income obtained, from the end user buyers, people with ready money, i.e. consumers and government. [However government spends, its purchases are always tax free!]
    In other words, taxes on production and services become costs that are indistinguishable from explicit consumer sales taxes, but which are largely hidden in retail prices. Everyone pays such an implicit sales tax at the same average general rate when buying goods and services with their real disposable income, i.e. after tax income.
    It is hard to deny that the complexity of the tax system is deliberately constructed to conceal the actual taxation of consumption, in order to support various political ideologies.
    Thus, when the collection of such sales taxes is avoided overseas, governments cry foul. The truth is that reverting to explicit sales taxes would solve the problem at home, but lead to uncomfortable questions from voters. So blame business!

Comments are closed.