Reflections on the recommendations of the Ethics Committee on the Guidelines for Observation and Exclusion of Companies from the Government Pension Fund Global
PAX works together with committed citizens and partners to protect civilians against acts of war, to end armed violence, and to build a just peace. We also strive to contribute to strong international standards and regulation concerning the protection of human rights.
PAX hosts several projects aimed at reducing humanitarian suffering caused by controversial weapons, including weapons of mass destruction. These efforts seek to reduce private sector investment into the producers of controversial weapons while increasing the understanding of socially responsible investment norms to preclude investment in inhumane weapons and redirect it towards sustainable societies. We also encourage investors to engage directly with weapons producers involved in the development of emerging technologies that might be linked to weapons, and the arms trade. We believe weapons should not be exported to states that violate international humanitarian law. To enable these efforts, we conduct primary research and engage in international advocacy work.
The activities of companies have enormous consequences and affect the lives of everyone — for better or for worse. Several actors have influence over company management and therefore company behaviour. These include institutional investors, such as the Government Pension Fund Global.
PAX submitted recommendations to the Council on Ethics regarding the Guidelines for Observation and Exclusion of Companies from the Government Pension Fund Global, we recommended the council consider the following:
· Adjust the methodology on nuclear weapon producer screening, and expand the exclusion list;
· Exclude companies known to be selling weapons fuelling the Yemen crisis;
· Add lethal autonomous weapon systems to its product-based exclusion list, and;
· Add meaningful human control as a principle for assessing the exclusion of weapon systems.
The report of the Committee (NOU 2020:7) of June 2020 included key recommendations to address these issues. In general, PAX supports the recommendations contained in the Committee report, and encourages their adoption for the reasons as outlined below.
PAX supports the recommendation to clarify the wording of the criterion to “ clarify that both weapons production and development are meant to fall within its scope, and that the criterion covers finished weapons and their key components alike .” Not only will this align the criterion wording with current practice, but also reinforces normative frameworks that recognise the development of inherently indiscriminate weapons is itself problematic.
Lethal Autonomous Weapons Systems
PAX welcomes the recommendation to add lethal autonomous weapons to the product-based exclusion list. This recognises that weapons which use sensor inputs to identify and apply force to targets without meaningful human control, inherently raise serious legal and ethical concerns. In looking at ways this recommendation will be interpreted, we suggest considering that systems that can change critical parameters based on machine learning, or become too complex for human users to understand and thus produce unpredictable and inexplicable effects are also of deep concern. To ensure meaningful human control a weapon system must be reliable, predictable, and understandable for the user, who must have sufficient information of the situation on the ground, the functioning of the weapon system and the interaction between the two. Technological advances incorporating machine learning and embracing automated decision making run the risk of increasingly complex systems and eliminate the option for meaningful human control.
The Committee noted that even if the “ future may see the emergence of technological sophistication capable of meeting the requirements of distinction under humanitarian law ” such weapons would remain ethically problematic. According to the International Committee of the Red Cross and Red Crescent, such weapons would also be legally problematic. Weapons themselves cannot be responsible for applying the law or be held accountable for violations; the law is addressed to human beings and they alone are responsible.
In particular, the implementation of this recommendation should ensure that the Fund avoids financial exposure to companies developing technology that would autonomously target humans, reducing people to data points. It is deeply unethical to delegate the decision over life and death to a machine and/or algorithms. This violates fundamental ethical principles, such as human dignity and the right to life. These issues were also raised in the 2015 Annual Report of the Ethics Council. As this technology is still under development, but progressing rapidly, the adoption of this recommendation enables early preventative dialogue with companies considering breaching this technical boundary.
With the pending entry into force of the Treaty on the Prohibition of Nuclear Weapons, any distinctions between nuclear and other conventional weapons begins to be erased. The recommendation to include key components in addition to finished weapons also provides the opportunity to treat the list of weapons considered to violate fundamental humanitarian principles through their normal use in an equitable manner. For example, while the Ethical Guidelines do not elaborate on nuclear weapons specifically, in 2004 the Revised National Budget included a list of the types of weapons the Fund shall not contribute to the production of, and specifically stated that the Fund shall not invest in companies that produce key components for nuclear weapons. [1] In applying this criteria to the Fund exclusions broadly, there are opportunities to refine screening methodology to prevent investment in companies currently producing key components for nuclear weapons, as was done with the expansion in 2018 of companies excluded by the fund for their association with the production of nuclear weapons. [2] This is a clear contribution to strengthening the norm against weapons of mass destruction.
As regards the recommendation to apply the exclusion criteria to certain types of delivery platforms, notably submarines, the development is welcome as it recognises the development of nuclear armed submarines requires specific technological capabilities and poses a proliferation risk. In implementing this recommendation, attention should be paid to articles described as ‘dual use’ (military and civilian) technology as it has been our experience that anything specifically designed for military purposes is always a military product at the outset, and only moves to a dual-use designation if at some point in time civilian uses of that specific good have become (more) prevalent (e.g. some sensor technologies).
PAX welcomes the Committee proposal to include “ a new conduct-based criterion for the sale of weapons to states involved in armed conflict where there is an unacceptable risk that the weapons are used in military operations that constitute serious and systematic violation of international humanitarian law .” However, in recommending a high threshold for application, the Committee does not consider that conclusive evidence linking specific weapons to specific violations is difficult to obtain. This would in effect make the threshold even higher, likely higher than intended. A high threshold would fail to consider the risk and could amount to post-fact exclusions only.
Conclusive evidence that specific weapons have been used in specific instances of attacks that violate principles of International Humanitarian Law is by definition difficult and rare. As weapons are produced to create destruction, by their very nature they leave little or no trace of evidence. Some reporting does exist that links companies to the deployment of specific weapon types. However, these reports are rare, and (logically) only focus on ammunition, not on delivery platforms such as aircraft or launch pads.
We therefore recommend that the Fund adopts an approach based on the risk that weapons will be used in violations. This risk is high (unacceptable) when an arms company supplies weapons systems to a state that has a record of committing human rights violations and violations of International Humanitarian Law. Instead of looking for direct links between the use of specific weapons, in specific violations, we recommend the Fund combine evidence of actual violations by states, with the supply of weapons systems. This would flag companies that supply weapons systems to states that commit violations and would thus protect the Fund from investments with a high risk of linkage to these violations.
A recent report by PAX, “ Controversial Arms Trade and investments of Dutch pension funds [3] ”, elaborates on the role of pension providers in the controversial arms trade and builds on conduct-based criteria. Whereas states party to the Arms Trade Treaty are meant to prevent sales of weapons to known human rights violators, institutional investors are well placed to reinforce these provisions. This report makes a strong case for most the arms companies in the report that they contribute to violations of international law in the war in Yemen, as their supplies of weapons continued after the war in Yemen started.
It is important to adopt this recommendation to incorporate ‘conduct-based’ analysis of weapons producers into the Fund’s criteria in order to prevent becoming linked to gross humanitarian harm and human rights violations. It would put these weapons companies on notice that their actions are unacceptable and cannot continue with impunity.
PAX welcomes the recommendations of the Committee towards the Council of Ethics and recognises the opportunities presented in the adoption and implementation of these adjustments. Continuing to centre the ethical obligation to achieve a good return and avoid unethical investments mean that this set of recommendations need to be able to be adapted to evolving norms and conditions.
PAX is eager to engage as these recommendations are adopted and the process of implementation creates new opportunities to prevent humanitarian harm.
Submitted by Susi Snyder, PAX, snyder@paxforpeace.nl
[1] Report No. 24 (2006-2007) to the Storting, available: https://www.regjeringen.no/contentassets/eeb49ccf49b94cfe881fdaa678efa537/en-gb/pdfs/stm200620070024000en_pdfs.pdf ;
[2] Norges Bank, “Observation and exclusion of companies”, 13 May 2020, available: https://www.nbim.no/en/responsibility/exclusion-of-companies/, viewed October 2020.
[3] The report can be accessed online: https://www.paxvoorvrede.nl/media/files/2019-11-praktijkonderzoek-controversial-arms-trade-and-investments-of-dutch-pension-funds.pdf
PAX works together with committed citizens and partners to protect civilians against acts of war, to end armed violence, and to build a just peace. We also strive to contribute to strong international standards and regulation concerning the protection of human rights.
PAX hosts several projects aimed at reducing humanitarian suffering caused by controversial weapons, including weapons of mass destruction. These efforts seek to reduce private sector investment into the producers of controversial weapons while increasing the understanding of socially responsible investment norms to preclude investment in inhumane weapons and redirect it towards sustainable societies. We also encourage investors to engage directly with weapons producers involved in the development of emerging technologies that might be linked to weapons, and the arms trade. We believe weapons should not be exported to states that violate international humanitarian law. To enable these efforts, we conduct primary research and engage in international advocacy work.
The activities of companies have enormous consequences and affect the lives of everyone — for better or for worse. Several actors have influence over company management and therefore company behaviour. These include institutional investors, such as the Government Pension Fund Global.
PAX submitted recommendations to the Council on Ethics regarding the Guidelines for Observation and Exclusion of Companies from the Government Pension Fund Global, we recommended the council consider the following:
· Adjust the methodology on nuclear weapon producer screening, and expand the exclusion list;
· Exclude companies known to be selling weapons fuelling the Yemen crisis;
· Add lethal autonomous weapon systems to its product-based exclusion list, and;
· Add meaningful human control as a principle for assessing the exclusion of weapon systems.
The report of the Committee (NOU 2020:7) of June 2020 included key recommendations to address these issues. In general, PAX supports the recommendations contained in the Committee report, and encourages their adoption for the reasons as outlined below.
PAX supports the recommendation to clarify the wording of the criterion to “ clarify that both weapons production and development are meant to fall within its scope, and that the criterion covers finished weapons and their key components alike .” Not only will this align the criterion wording with current practice, but also reinforces normative frameworks that recognise the development of inherently indiscriminate weapons is itself problematic.
Lethal Autonomous Weapons Systems
PAX welcomes the recommendation to add lethal autonomous weapons to the product-based exclusion list. This recognises that weapons which use sensor inputs to identify and apply force to targets without meaningful human control, inherently raise serious legal and ethical concerns. In looking at ways this recommendation will be interpreted, we suggest considering that systems that can change critical parameters based on machine learning, or become too complex for human users to understand and thus produce unpredictable and inexplicable effects are also of deep concern. To ensure meaningful human control a weapon system must be reliable, predictable, and understandable for the user, who must have sufficient information of the situation on the ground, the functioning of the weapon system and the interaction between the two. Technological advances incorporating machine learning and embracing automated decision making run the risk of increasingly complex systems and eliminate the option for meaningful human control.
The Committee noted that even if the “ future may see the emergence of technological sophistication capable of meeting the requirements of distinction under humanitarian law ” such weapons would remain ethically problematic. According to the International Committee of the Red Cross and Red Crescent, such weapons would also be legally problematic. Weapons themselves cannot be responsible for applying the law or be held accountable for violations; the law is addressed to human beings and they alone are responsible.
In particular, the implementation of this recommendation should ensure that the Fund avoids financial exposure to companies developing technology that would autonomously target humans, reducing people to data points. It is deeply unethical to delegate the decision over life and death to a machine and/or algorithms. This violates fundamental ethical principles, such as human dignity and the right to life. These issues were also raised in the 2015 Annual Report of the Ethics Council. As this technology is still under development, but progressing rapidly, the adoption of this recommendation enables early preventative dialogue with companies considering breaching this technical boundary.
With the pending entry into force of the Treaty on the Prohibition of Nuclear Weapons, any distinctions between nuclear and other conventional weapons begins to be erased. The recommendation to include key components in addition to finished weapons also provides the opportunity to treat the list of weapons considered to violate fundamental humanitarian principles through their normal use in an equitable manner. For example, while the Ethical Guidelines do not elaborate on nuclear weapons specifically, in 2004 the Revised National Budget included a list of the types of weapons the Fund shall not contribute to the production of, and specifically stated that the Fund shall not invest in companies that produce key components for nuclear weapons. [1] In applying this criteria to the Fund exclusions broadly, there are opportunities to refine screening methodology to prevent investment in companies currently producing key components for nuclear weapons, as was done with the expansion in 2018 of companies excluded by the fund for their association with the production of nuclear weapons. [2] This is a clear contribution to strengthening the norm against weapons of mass destruction.
As regards the recommendation to apply the exclusion criteria to certain types of delivery platforms, notably submarines, the development is welcome as it recognises the development of nuclear armed submarines requires specific technological capabilities and poses a proliferation risk. In implementing this recommendation, attention should be paid to articles described as ‘dual use’ (military and civilian) technology as it has been our experience that anything specifically designed for military purposes is always a military product at the outset, and only moves to a dual-use designation if at some point in time civilian uses of that specific good have become (more) prevalent (e.g. some sensor technologies).
PAX welcomes the Committee proposal to include “ a new conduct-based criterion for the sale of weapons to states involved in armed conflict where there is an unacceptable risk that the weapons are used in military operations that constitute serious and systematic violation of international humanitarian law .” However, in recommending a high threshold for application, the Committee does not consider that conclusive evidence linking specific weapons to specific violations is difficult to obtain. This would in effect make the threshold even higher, likely higher than intended. A high threshold would fail to consider the risk and could amount to post-fact exclusions only.
Conclusive evidence that specific weapons have been used in specific instances of attacks that violate principles of International Humanitarian Law is by definition difficult and rare. As weapons are produced to create destruction, by their very nature they leave little or no trace of evidence. Some reporting does exist that links companies to the deployment of specific weapon types. However, these reports are rare, and (logically) only focus on ammunition, not on delivery platforms such as aircraft or launch pads.
We therefore recommend that the Fund adopts an approach based on the risk that weapons will be used in violations. This risk is high (unacceptable) when an arms company supplies weapons systems to a state that has a record of committing human rights violations and violations of International Humanitarian Law. Instead of looking for direct links between the use of specific weapons, in specific violations, we recommend the Fund combine evidence of actual violations by states, with the supply of weapons systems. This would flag companies that supply weapons systems to states that commit violations and would thus protect the Fund from investments with a high risk of linkage to these violations.
A recent report by PAX, “ Controversial Arms Trade and investments of Dutch pension funds [3] ”, elaborates on the role of pension providers in the controversial arms trade and builds on conduct-based criteria. Whereas states party to the Arms Trade Treaty are meant to prevent sales of weapons to known human rights violators, institutional investors are well placed to reinforce these provisions. This report makes a strong case for most the arms companies in the report that they contribute to violations of international law in the war in Yemen, as their supplies of weapons continued after the war in Yemen started.
It is important to adopt this recommendation to incorporate ‘conduct-based’ analysis of weapons producers into the Fund’s criteria in order to prevent becoming linked to gross humanitarian harm and human rights violations. It would put these weapons companies on notice that their actions are unacceptable and cannot continue with impunity.
PAX welcomes the recommendations of the Committee towards the Council of Ethics and recognises the opportunities presented in the adoption and implementation of these adjustments. Continuing to centre the ethical obligation to achieve a good return and avoid unethical investments mean that this set of recommendations need to be able to be adapted to evolving norms and conditions.
PAX is eager to engage as these recommendations are adopted and the process of implementation creates new opportunities to prevent humanitarian harm.
Submitted by Susi Snyder, PAX, snyder@paxforpeace.nl
[1] Report No. 24 (2006-2007) to the Storting, available: https://www.regjeringen.no/contentassets/eeb49ccf49b94cfe881fdaa678efa537/en-gb/pdfs/stm200620070024000en_pdfs.pdf ;
[2] Norges Bank, “Observation and exclusion of companies”, 13 May 2020, available: https://www.nbim.no/en/responsibility/exclusion-of-companies/, viewed October 2020.
[3] The report can be accessed online: https://www.paxvoorvrede.nl/media/files/2019-11-praktijkonderzoek-controversial-arms-trade-and-investments-of-dutch-pension-funds.pdf