This is the second blog post in a series looking at an array of issues in 2018 related to weapons use, the arms trade and security assistance, at times offering recommendations.
Sweden, as part of the European Union, is already bound by the EU Common Position on arms exports, which requires an export license to be denied if, among other things, there is a clear risk that the equipment to be exported might be used for violations of human rights or international humanitarian law. However, this is always part of a case-by-case evaluation relating to the specific equipment being sold; the status of the recipient state as such plays no direct role in the criteria, except in the small number of cases where there is either a UN or an EU arms embargo. The democracy criterion would therefore be significantly stricter than the EU Common Position—and indeed most if not all major arms exporters.
In fact, Swedish law already goes beyond the Common Position in terms of considering the status of the recipient, in relation to human rights abuses, but the new law clearly strengthens this.
Specifically, the law requires that the proposed recipient state’s democratic status—relating to the existence of democratic institutions, the possibility for freedom of expression and respect for fundamental democratic principles—shall be a “central condition” in the evaluation of export license applications, and that serious failings in these will constitute an “obstacle” to approval. Regarding human rights, where the existing law states that “gross and systematic” violations of human rights constitute an obstacle to approving a license, the bar is lowered to “serious and systematic” abuses. The potential for a sale to counteract sustainable development is also to be considered, and there are further measures to improve transparency in arms sales.
What the law does not enact, however, is an absolute ban on arms exports to dictatorships or to human rights abusers; each export license application is still to be judged individually, based on an overall judgment taking into account numerous factors, including those relating to security and defense policy. Thus, arms sales to non-democratic states could still be approved if the government (through the export control agency, ISP, the Inspectorate for Strategic Products) decides that defense industrial considerations outweigh the recipient’s lack of democracy. As far as I can tell, the law does not specify exactly how these issues are to be weighed against one another.
This lack of a clear ban on arms sales to non-democratic regimes, even dictatorships, has led to criticism from the peace movement, and in Parliament by the Left Party. Svenska Freds (Swedish Peace) has been particularly critical of the proposed law as leaving open too many loopholes to allow, essentially, business as usual. The proposal states that lack of democracy will be an “obstacle” to license approval, but, asks Svenska Freds Chairperson Agnes Hellström, “How high an obstacle”?
Other parties, including the Green Party and the generally right-wing Christian Democrats stated that they would have preferred a complete ban on arms sales to dictatorships, but accepted the cross-party proposal as a reasonable compromise. The two largest parties in the Swedish Parliament, the Social Democrats and the center-right Moderates both have strong traditional ties to the arms industry (via the unions and the business community respectively), and the former were the architects of the post-War policy of armed neutrality under which the Swedish arms industry—still remarkably advanced for a country of 10 million—was developed. Thus, a more absolute measure would have been unlikely to gain favor with these dominant players.
The question that will only be answered with time is, will the new regulations lead to a significant reduction in arms sales to undemocratic regimes, and perhaps a complete halt to sales to the worst dictatorships, or will the new form of words be simply used by the government to claim that it has fulfilled the promise of a ”democracy criterion,” while continuing with business as usual? (Also to be seen is whether the interests of sustaining an arms industry facing a severely limited domestic market will always be high enough to vault the democracy and human rights obstacles.)
A starting point for judging the new law is examining how Sweden’s and other major European exporters’ export control regimes are already willing to export to repressive regimes. (Please see note on data further below)
Share of exports to “Not-Free” countries
Country Deliveries 2007-16 Licenses 2012-16
United Kingdom 49% 38%
France 37% 51% (orders)
Germany 12% 36%
Italy 26% 15%
Spain 19% 13%
Sweden 15% 15%.
Sweden clearly comes toward the lower end on measures of deliveries (2007-2016) and recent licenses (2012-2016), with similar levels to Spain, although Germany has a lower share of deliveries over the longer period 2007-16. For licenses, the great bulk of Sweden’s export licenses to dictatorships come from a single $1.3 billion deal in 2016 to sell Saab Erieye Airborne Early Warning and Control systems to UAE—of particular concern given the UAE’s role in the air war against Yemen, where such advanced sensor systems could play a significant role. Sweden also sold an earlier version of the Erieye to Saudi Arabia in 2010 (delivered 2014). Other significant sales to “Not Free” countries include a 2012 sale of missiles and sensors to Algeria.
One of the most controversial deals in Sweden in recent years, however, was the sale in 2008 and then 2010 of Gripen multi-role combat aircraft to Thailand. The deal was negotiated under the military regime that took power in Thailand through the coup of 2006, but the orders and deliveries took place between the restoration of civilian rule following the general election of December 2007, and the most recent coup of 2014. Nonetheless, the sale of such major equipment to a military establishment with a long history of coups, and where the return democracy was always highly fragile, was controversial. In the end, Sweden approved export licenses for delivery of around $100 million worth of military equipment to Thailand—now firmly in the grip of military rule—in 2016.
While exports to Saudi Arabia have been a major focus of concern, in fact since the Erieye deal, export license approvals to the Saudis have been negligible; partly due to the breakdown in Swedish-Saudi relations following the non-renewal by Sweden in 2015 of a military cooperation agreement (although this agreement had produced few if any tangible results), but very likely also in part due to the high sensitivity of such sales. From 2014-16, license approvals to Saudi have amounted to just a few million crowns (less than $1 million).
Thus, while Sweden does sell arms to dictatorships and human rights abusers, the evidence suggests that, in comparison with most other significant European arms producers, concerns over human rights do lead to some degree of restraint on the part of Swedish authorities. The new rules certainly have loopholes that will allow the Swedish government to sell arms to dictatorships when it really wants to, and it is possible that there will be no practical change; however, it seems likely that the strong level of pressure for restraint from both civil society, media and elements within Parliament, and the expectations of such accompanying the new law, will lead to some greater degree of caution on the part of ISP. For example, any significant licensing of arms sales to Saudi Arabia would immediately draw loud outcries and would (not unreasonably) be taken as proof that the new law is in practice meaningless. Sales to Saudi also jar heavily with Sweden’s declared feminist foreign policy. But a complete end to arms sales to undemocratic regimes is not on the cards.
Either way, assuming the new law is passed, Sweden will be a country worth watching to see if public opprobrium against the arms trade can create sufficient political will to make a more restrictive arms export policy a reality, in the face of the strong defense industrial pressure to maintain exports.
Samuel Perlo-Freeman is a Fellow at the World Peace Foundation (WPF), and Project Manager for the WPF project on Global Arms Business and Corruption
Notes on the data:
Analysis looked at the top 6 EU arms industries--the UK, France, Germany, Italy, Spain, and Sweden-- comparing across two measures: first, SIPRI data for deliveries of major conventional weapons over the period 2007-16, using their Trend Indicator Value (TIV) volume measure; second, the value of export licenses granted over the period 2012-16. For license values, unfortunately France changed their export licensing system in 2015 in a way that led to a massive increase in their license values, in a way that makes the data almost meaningless. Therefore, I used figures for export orders instead, which only France provides.
For each measure, I looked at the total value of exports, and within this the share of the total going to countries classified as “Not Free” by Freedom House.
There are two caveat to the SIPRI TIV measure: first, it is a volume rather than a financial measure; since Western European countries have similar levels of income, unit costs are probably comparable, so this issue is not critical. Secondly, SIPRI data only includes major conventional weapons, which include some subsystems such as radars and engines, but not most. This tends to understate UK exports in particular, where a large proportion involves subsystems and components.
There are likewise caveats to the license data: first, not all licenses turn into deliveries, and the ratio of licenses to deliveries over time varies considerably between countries. I have not checked if this varies also by recipient country. Second, export license data only covers individual licenses, which allow a single delivery of specified equipment. It does not include open or general licenses, which allow for multiple deliveries of a category of equipment, and which therefore do not carry a financial value. The UK makes particularly high use of open licenses, so that export license figures severely understate the value of UK exports. Again, whether this understatement differs significantly by type of recipient, and therefore may distort the shares below, is unclear. The value of license data is that it gives the most direct idea of export license decision-making by governments.
 SIPRI Arms Transfers Database
 Swedish arms export control report for 2016.
 Swedish arms export control reports.
 License and order values come from national reports on arms export controls, available from SIPRI at https://www.sipri.org/databases/national-reports; export for the UK, where data is taken from the Campaign Against Arms Trade database on UK arms export licenses, which presents data in a far more user-friendly fashion than government sources; see https://www.caat.org.uk/resources/export-licences.