New protocol to the Sweden-Austria tax treaty on capital gains taxation.

More on the Sweden-Austria tax treaty

(by Peter Sundgren)

The Sweden-Austria tax treaty which was signed originally in1959 is the oldest bilateral tax  treaty in force as far as both countries are concerned. At its latest revision in 1991, it was agreed that the treaty was outdated in several respects and that one should negotiate a new treaty “as soon as possible”. Talks were subsequently started and resulted in an initialed draft in 1995. Since then, however, nothing has happened. When inquiring about the treaty the answer, over the last eleven years, has been that an “exchange of letters is in progress”.

Article 8 of the treaty gives exclusive taxing rights on share gains to the country of residence. In addition Austria since 1 January 1992, only taxes that part of the appreciation of the shares that occurs during the shareholders residence in Austria. A common description hereof is that Austria ‘steps up’ the acquisition cost of the shares to the market value at the time of entry of the tax payer to Austria. Accordingly, a Swedish resident shareholder who moves to and becomes a resident of Austria  will be exempted from capital gains taxation in Sweden under the treaty and pay Austrian tax only on that part of the appreciation that takes place after the shareholder moved to Austria. If the sale of the shares takes place immediately thereafter the gain will of course be (virtually) nil. Or, in other words, the interaction of the treaty and Austrian domestic tax law gives rise to a double non-taxation situation. As aforementioned this situation has now prevailed for fifteen years without anything having been done about it.

Interesting in a still wider historical context is that Sweden in 1983 introduced a so called 10-year rule subjecting Swedish residents who move abroad to Swedish capital gains taxation if the sale occurs at any time during a ten year period after giving up Swedish tax residence. Parliament of course realized that this law, with regard to treaty states where the capital gains tax article was drawn up as it is with for instance Austria, would be quite ineffective and therefore instructed the Government to go ahead and renegotiate these treaties. This is now twenty-three years ago. According to a recent Austrian notice Sweden has indeed attempted to put a stop to the present “Steuerschlupfloch” in the treaty for the last ten years but that these efforts have been “dilatorily” treated by the Austrians.

Several times in the 90s I reported from the Institute of Foreign Law about the developments of the Sweden-Austria treaty and in 2001 I concluded that the procrastination, however, “was good news for Swedish shareholders moving to Austria who could thus continue to collect their tax free share gains”.

In late 2005 I gave a speech at the Gothenburg Business School reporting that the swell of Swedish shareholders moving to Austria had risen to tsunami proportions. In February of this year I wrote a notice on the matter on this webbsite.

A couple of weeks later  our (social democratic and now ex-)finance minister Pär Nuder responded to a question on a radio interview why nothing had been done about the capital gains ‘loop-hole’ in the Sweden-Austria tax treaty. Because he had been in office for such a short time he unblushingly blamed his predecessors herefor. Then, however, he sheepishly added that he would immediately look into the matter.

In the meantime, the tax authorities have been engaged in investigating the extent of the tax losses resulting from this double non-taxation situation. Their conclusion is mind boggling! Accordingly, it has been determined that in the time period 2000-2005, the shortfall of Swedish capital gains taxes has amounted to 3.1 billion (3100 000 000 million) SEK. An Austrian estimate of that period is 500 million Euros. And this does not take into consideration the big ‘tax kills’ that took place during the IT-boom in the late 90s. Therefore, all told, it seems like the loss to the Swedish revenue from this little ‘Austrian affair’ alone may have amounted to around 8-10 billion SEK. This compares to  about a quarter of the Swedish defence budget!

Now things began to move fast in the Finance Ministry – very fast indeed! Urgent talks were resumed with Austria in May and a new protocol to the treaty was drafted by early July. This text was immediately, and in the midst of our (lazy) summer vacation season, remitted to the Regional Tax Appeals Court (Kammarrätten) in Jönköping and the Tax Authorities for scrutiny who both rapidly reported back that they had no objections regarding the protocol. Shortly thereafter, on 21 August, the protocol was signed by the Swedish and Austrian Governments.

Then things started to go awry.

But first a short account of the protocol: It allows Sweden to tax that part of the gain on the shares that has accumulated up and until the time in point that the shareholder becomes a resident of Austria and the rest of the gain, in accordance with its legislation, may be taxed by this latter country. This method, whereby the tax base is divided between the two contracting states is quite extraordinary. (It has,however, existed in a terminated treaty with New Zealand and the income of the Kirunavaara-Lousavaara minefields in Lappland, under a special agreement, is also shared by Sweden and Norway.)

A problem, however, regarding the new protocol, has cropped up in its enter into force provisions! These rules provide:

a)      that the protocol applies only to individuals that have taken up residence in Austria later than 21 August when the protocol was signed, (so that unsuspecting and innocent shareholders who moved to Austria before that date would not be affected),

b)      that the protocol enters into force on the thirtieth day after the day when appropriate notices have been exchanged determining that the two states respectively have completed the constitutional measures required for making the protocol enter into force and

c)      that the new rules shall apply only to sales occurring on 1 January 2007 or later.

The intention has no doubt been that the measures under b) above would take place during 2006. As it now appears, however,  Austria may not be able to complete its ratification process in that time but only later, sometime in 2007. This would then mean that the a sale under c) above may occur before the protocol has entered into force according to b). And this may constitute a breach of our constitutional ban against retroactive legislation!

With a speed that has seldom been experienced before, (and certainly never as far as a tax treaty protocol is concerned), the finance ministry has drafted the Government bill on the protocol, remitted it again to the Regional Tax Court in Jönköping and the Tax Authorities and this time also to the Stockholm Court of (civil and criminal) Appeal (!), (Svea Hovrätt), The Confederation of Swedish Enterprise (Svensk Näringsliv) and The Federation of Private Enterprises (Företagarna).

The Jönköping Court has suggested that the exceptional provision under chapter 2 , § 10 of our Constitution allowing retroactive legislation in certain emergency cases may apply, but advises that this be subjected to a thorough investigation. The Federation of Swedish Enterprise rejects the applicability of this exceptional provision in the Constitution because no such tax avoidance required under that provision has taken place.

The replies from these institutions have also been very prompt and after government approval on 14 September of the new Austrian protocol bill it has been sent to the Swedish Law Council (Lagrådet) for their opinion on the constitutional issue just described. This kind of remittance to the Law Council, too, has never occurred in the history of tax treaties! (Consequently, if one is critical of the government’s somnambulant handling of the Sweden-Austria situation over the last couple of decades one must on the other hand admire the eruption of activity they have demonstrated in the last four months (even if that has taken place, as we say in Sweden, ‘under the gallows’)!

In the meantime, according to rumors, there is a stampede of Swedish entrepreneurs now heading for Austria in order to squeeze through the enter-into-force ‘key-hole’ of the protocol. One can expect that the tax authorities will investigate very closely the time-point of the taking up of their tax residence in Austria and the dating of their sales contracts.

Experience has shown that emigrating shareholders in many cases will return to Sweden after only a short interval evidencing that the transfer of their residence abroad was undertaken essentially in order to sell the shares tax-free under the domestic rules in the new country of residence and escape taxation in Sweden, This problem is indeed highlighted in the OECD Model Treaty’s discussion on improper use of conventions, section 9 of the Commentaries to Article 1. Considering the problems of counteracting these sham constructions in practice, The Federation of Swedish Industries, a couple of years ago, in their comments to a government report regarding our domestic residence regime on individuals, suggested that one should introduce a so called ‘claw-back’ rule for emigrating Swedish residents, allowing Sweden to tax share sales, if the shareholder returned to Sweden within, say, five years and during that time abroad had sold any shares he had owned when he left Sweden. The UK has such a regime.

This Sweden-Austria debacle gives support to my long held view that tax treaties to a very large extent are exploited for unwarranted reasons and not used as instruments for avoiding double taxation. Such double taxation, at least as far as Sweden is concerned, is a phenomenon of the past. Reduced source taxes under EU-provisions and other agreements etc. together with effective domestic Swedish relief of foreign tax measures have to a large extent made our treaties obsolete for this purpose. Tax practitioners today exploring our tax treaties are making no money advising people how to avoid double taxation. Their main occupation is to exploit these treaties so that taxation is avoided altogether. 

As mentioned above a draft treaty was negotiated with Austria already in 1995. This draft is not open for inspection but it would be highly improbable that it did not give Sweden primary rights to tax share gains according to our ten year rule. Therefore, what must be considered as nothing but sheer indolence and inefficiency over a very long time period  leading to the squandering of tax payer money on an unprecedented scale represents an enduring embarrassment to the Swedish government. A government, which has always prized itself for its tough stand on tax avoidance! This deplorable affair has effectively demonstrated the emptiness of that rhetoric.

It is time for our tax agencies to wake up to the perils and scope of  the unacceptable international tax planning and tax avoidance that is going on in our country.(Especially in the field of capital gains taxation of individuals. Many previous ‘treaty affairs’ also bear testimony hereto.)

In order to prevent scandals of this type in the future one should consider setting up a ‘watch-dog’ function to monitor the development of the application of our tax treaties. Taking into account that measures against international tax avoidance require a deep knowledge of the tax systems of foreign jurisdictions one could even consider assigning this task to tax experts in those countries, in particular in our most important treaty partner states. (Such experts could also be helpful when negotiating new treaties.) In addition, Competent Authorities should take more seriously their obligation under tax treaties (usually in article 2 thereof) to notify each other of any significant changes that have been made in their taxation laws. Exchange of information and cooperation between authorities must also be stepped up.

peter@sundgren.net