Day: November 20, 2022

Matt Boyd, Nick Wilson, Ben Payne

(12 min read)

 

TLDR:

• The draft NZ National Security Long-term Insights Briefing (already informed by a public survey) is currently out for public consultation.
• The briefing identifies national security trends, potential scenarios, and provides additional detail on six important national security issues: Disinformation, Hacking and cyber attacks, Transnational organised crime, Foreign interference and espionage, Terrorism and violent extremism, Pacific resilience challenges.
• The briefing identifies ten features that could help progress to a secure future for New Zealanders.
• This post details our three suggestions to improve the briefing: (1) important improvements to future iterations of the public survey, (2) the need to explicitly articulate the extreme tail risks of each major trend identified, and (3) the importance of signalling a move towards an integrated and comprehensive National Risks Assessment.

The Draft NZ National Security LTIB (November 2022)

The draft NZ National Security Long-term Insights Briefing (LTIB) has been presented for public consultation.

The LTIB was produced by a group of nine government agencies that are responsible for protecting NZ from national security threats.

The briefing draws on information obtained from communities, businesses, and government agencies. This includes information from a survey, which we discuss below.

The LTIB examines key global trends across the next 10 to 15 years, some national security risks, and challenges ‘of concern’ and how they might change across time.

Four key global trends are identified:

• Increasing competition and continued deterioration of the international ‘rules-based order’
• Transformational technology changes
• Climate change
• Covid-19 and future pandemics

Three plausible global scenarios are outlined:

• Continued decline: continuing armed conflict, competition for resources and the impact of malicious actors
• Dramatic decline: spreading conflict, unmitigated climate impacts and the possibility that ‘a nuclear weapon’ could be used
• Optimistic scenario: including international collaboration, technological innovation, and investment in climate adaptation.  

The draft of the National Security LTIB focuses on six security issues:

• Disinformation
• Hacking and cyber attacks
• Transnational organised crime
• Foreign interference and espionage
• Terrorism and violent extremism
• Pacific resilience challenges

Each threat is profiled and followed by sections describing ‘what we expect to see in the next 10-15 years,’ and how NZ can be ‘preparing for the future now’.

The LTIB rightly acknowledges the competing demands of, ‘investing in response to current crises and building our capacity and capability to respond to future challenges, including preparing for high impact but rare events.’

The briefing concludes with Ten Features that could support a bright future for national security. 

• Transparent accessible public information
• National security sector stewardship
• Strengthened political leadership on national security
• International partnerships that grow and strengthen our national security
• A national security sector that reflects the diversity of our nation and is adaptable and capable of responding to future challenges
• Open debate among, and advice from, experts outside of government
• Active and engaging media coverage
• Recognising and working with partners outside of government
• Communities that feel enabled and empowered to engage with the national security sector
• Trusted and accountable institutions

Our feedback on the draft LTIB

The NZ Government and its officials are to be congratulated on this excellent draft National Security LTIB. This work represents a substantial move forward in thinking in this domain. Nevertheless, we take this consultation opportunity to offer ideas for further developing this document and for its future iterations – as outlined below.

1. Scope for Improving Subsequent Surveys of the Public

The survey of public opinion conducted by Ipsos provides a lot of interesting information. We understand that the survey questions were chosen to be consistent with global research undertaken by Ipsos. However, as a basis for decision-making the survey has some shortcomings which should be addressed in future iterations or through other public engagement channels.

Survey results show that New Zealanders want more information about the threats the country faces. The LTIB could specifically advocate for additional research and knowledge generation/dissemination about risks and threats that are poorly understood. The survey also revealed that the NZ population wants the opportunity to comment on national security issues (we discuss this below).

Potentially fruitful areas for improvement in the next iteration of the survey are as follows:

Firstly, many survey respondents indicated that they wanted more information about national security issues. This begs the question of whether their responses are fully informed responses or merely guesstimates based on partial information.

Secondly, the descriptions of some threats are either too broad or too specific.

• For example, one of the items included was ‘nuclear, chemical or biological attack somewhere in the world’. The use of a single nuclear weapon could escalate into nuclear war, the effects of which NZ would not escape (see our recent study here: [1]; and previous NZ work on nuclear war impacts here: [2] [3] [4] [5]). However, it was not clear whether respondents were to consider threats such as an all-out NATO-Russia nuclear war and any ensuing global nuclear winter, or whether respondents were contemplating something as localised as chemical attacks in the Syrian conflict. The potential impacts are vastly different – since some nuclear war scenarios could potentially result in the permanent collapse of civilisation.

• Some threats as described were possibly overly specific. For example, ‘breakdown in national infrastructure due to attack’ might have implied a direct cyber or physical assault on infrastructure. However, a breakdown in national infrastructure could occur due to the cascading impacts of other threats, such as Northern Hemisphere conflict and abrupt NZ trade isolation, or the impacts of a large supervolcanic eruption, devastating solar flare, or nuclear war. The more causal paths leading to an effect, the more probable (and threatening) a scenario might be ranked by respondents.

• This problem of ‘scenario choice’ has important implications for public engagement on risk and we discuss it in our recent paper on national risk assessments and national risk registers [6].

Thirdly, the importance of a threat is often taken to be some product of two factors, probability and consequences, which together generate an estimate of consequences in expectation.

• Some survey questions clearly asked about consequences, ‘IF these were to happen… which do you think poses the greatest threat’. However, other questions used ambiguous phrasing, ‘how real do you feel the threat is of any of the following happening…’ It is unclear whether respondents would interpret this as asking about probability, consequences, a combination of probability and consequences, or something else. It is particularly ambiguous given that the word ‘threat’ is also used in the question about consequences. If interpreted as probability by respondents then the latter question reveals nothing about the salience or importance of a risk, this could only be deduced by combining responses to both questions. However, when presenting the following matrix of government capability vs ‘level of perceived threat’ it is the ambiguous item that is graphed. The problem is that a respondent might feel there is a high probability of something happening, but that the consequences are trivial. In which case ‘Act and Improve’ (see ‘Survey Report’ p.10) is the wrong response. There is therefore a strong case for a future survey that accesses more decision relevant risk information across a more comprehensive spectrum of risks. The LTIB should ideally indicate aspiration to commission or conduct such survey work.

• The survey might also imply (if we interpret ‘how real do you feel the threat is of any of the following happening’ as accessing consequences in expectation, not merely probability) that violent conflict in NZ, a nuclear ‘attack’, or breakdown of infrastructure in NZ are more salient (deserve more focus moving forward) than, for example, terrorism and violent extremism, since the respondents felt that the government was already well placed to deal with the latter threats. Importantly nuclear attack (the LTIB should be far more explicit that major nuclear war is possible, not merely an ‘attack’), breakdown of infrastructure and violent conflict in NZ are not independent risks.

Future iterations of the survey could be conducted with improved wording, after a public information campaign, and include the full spectrum of risks.

We have previously argued for a two-way communication platform connecting information gathered through a comprehensive National Risk Assessment (all risks, not just national security, see below) with a portal for ongoing public and expert feedback and scrutiny [6]. A comprehensive LTIB could signal aspirations to develop such a tool, promoted by the media, and thereby effect the goals of the previous paragraph as well as advance the desirable national security features of:

• Transparent accessible public information
• Open debate among, and advice from, experts outside of government
• Active and engaging media coverage
• Recognising and working with partners outside of government
• Communities that feel enabled and empowered to engage with the national security sector

2. Suggestions for completing the Draft National Security LTIB

We agree with the four trends outlined in the LTIB. Increasing global competition, technological advance, climate change and pandemics are growing threats. But each of these four trends has an associated global catastrophic or existential threat to humanity and these extreme tail risks should be acknowledged and analysed because they would be unbearable.

Those that study the largest risks the world faces often prioritise risks from nuclear war, artificial intelligence not aligned to human values, extreme climate change, and engineered pandemics (eg, well summarised in work by Ord [7]). These four risks correspond with the four trends outlined. However, the wording of the LTIB could be clarified to specify that each trend has a globally catastrophic form, or even a tail risk that is an existential threat to humanity. NZ is in a privileged position in that it may well suffer less direct consequences from some of these threats (nuclear war [1], supervolcanic disaster [8]), however as our research has indicated, NZ may be extremely fragile to the cascading consequences of such major threats [1].

Looking to the results of the survey, of these four trends, ‘health epidemic’ tops the list of NZ citizens’ concerns (however they are interpreted). This is probably appropriate, especially if including concerns around natural pandemics (ongoing Covid-19 harms and newly emerging pandemic diseases), bioweapon pandemics, laboratory accidents, and gain-of-function engineering. Nevertheless, the LTIB could be more explicit that the scope of pandemic threats is not limited to influenza pandemics or Covid-19-like events and future biological threats could be extreme (even being existential threats to humanity [7]).

It is particularly appropriate to include misinformation as a major national security risk. A healthy information environment that facilitates accurately informed public discussion on threats is essential, and underpins the ability to analyse, prevent, respond to, and recover from all other threats.

Although the LTIB profiles five important threats other than misinformation, the omissions are interesting. It is not entirely clear why some threats depicted in the four-quadrant figure (public survey results p.10) were selected for inclusion and others for exclusion from further discussion in the LTIB.

• For example, domestic terrorism is rated by survey respondents as a moderate threat, that is well-handled by the government at present. Whereas nuclear/biological/chemical attack is rated as at least as threatening, but poorly addressed by the government (noting the ambiguity we discuss above over probability, consequences, and consequences in expectation in the survey).
• Some emphasis could be placed on unknown threats. There are likely many risks both internal and external to NZ and indeed to government itself that we do not know about yet (eg, pandemics from synthetic bioweapons was not appreciated as a future threat before genetic engineering technology was invented). General resilience building may be able to mitigate the impact of such unknowns.

The LTIB acknowledges the existence of tension between prioritising resources for current risks (eg, misinformation) vs future risks (eg, nuclear war). This tension is a strong argument for ensuring a methodology to systematically analyse national risks (present and future, national security risks and other risks) in terms of the ‘level of threat’ and how ‘concerning’ they are (with appropriate operationalisation of these concepts, making clear what kinds of wellbeing, assets, and values – now or in the future – are being considered) and make both the methodology and findings of the analysis public, so the public can have a say in resource prioritisation decisions.

Finally, although ‘natural disaster in New Zealand’ could be considered a national security threat, or at least a national risk, then relevant ‘natural disasters outside NZ’ should be included, such as supervolcanic eruptions or coronal mass ejections (solar flares). However, we suspect that an information campaign would be needed to obtain useful public engagement on such risks (given the technical complexities).

The LTIB states that, ‘New Zealand’s approach to building resilience in our society and preventing the spread of disinformation needs to be comprehensive and long term.’ We concur and this should be the case for every threat ‘of concern’. It is also relevant to particularly catastrophic threats.  For example, nuclear conflict that rises well beyond the use of ‘a nuclear weapon’ and leads to a possible nuclear winter. A multi-decade (long-term!) strategic plan to build NZ’s resilience could be envisioned as an incremental ongoing project. This could be seen as a priority given NZ’s privileged (high-income remote island nation), though fragile, position in greater context of humanity.

Unless the probability of a risk is zero then it will (by definition) occur given a long-enough timeframe. If the probability is one percent per annum (as is plausible with several global catastrophes eg, nuclear war [9]) then it is likely to happen this century if preventive measures are not scaled up. If each successive LTIB only looks 10 years ahead, then each may be blind to such risks. Some generation at some point must prepare for these risks, or all will suffer.

3. The National Security LTIB should signal a move towards a comprehensive National Risk Assessment

Analysis of national security threats needs to feed (along with other risk issues, eg, natural hazards, existential risks) into a National Risk Register that is aligned with a National Risk Mitigation Strategy & Plan, which includes a methodology for prioritising resources to the greatest threats.

The LTIB takes the perspective that national security issues are by and large agential threats, ie, those where someone or some group poses a threat. Other entities in NZ allegedly deal with other kinds of threats, such as natural hazards and other non-agential threats. However, this arrangement risks leading to siloed focus on historical threats with the result that risks that are less familiar to policymakers, emerging risks, and interacting risks slipping between the gaps. How these entities nest and communicate is very important, because as was highlighted in the UNDRR Framework for Global Science In Support of Risk-informed Sustainable Development and Planetary Health (2021) siloing is paralysing risk mitigation action across the entire spectrum of risks (agential and non-agential)

We think that risk management at a national level in NZ would benefit from a move towards an integrated risk approach. Stewardship and partnerships, as indicated in the ‘ten features’ will be important, but these partnerships should include those outside traditional ‘security’ silos, and there should be an overarching entity accountable for analysis, communication, and resource allocation recommendations across all national risks. This entity could be a Parliamentary Commissioner for national/extreme risks [10], or some other office.  

Part of the argument for the national security LTIB signalling the merits of moving to an integrated risk approach rests on global trends towards such approaches. We note that many countries publish a National Risk Register that includes both natural and agential risks. We also note that the distinction between natural and human-induced is often vague. International frameworks such as the Sendai Framework for Disaster Risk Reduction are moving away from such a distinction, with the UN General Assembly deciding to hold a ‘midterm review of the implementation of the Sendai Framework 2015-2030.’ As part of this process, the Global Platform on Disaster Risk Reduction 2022 held a plenary session on ‘beyond natural hazards – operationalizing expanded scope’, and argued in favour of the need to take into account anthropogenic risks and take more preventive action.

The future will likely see the Sendai Framework integrating risks traditionally seen as ‘natural’ and threats perhaps traditionally seen as ‘security’ issues. The plenary session noted that there was a risk of being, ‘blind to the full range of global catastrophic or existential risk scenarios’.

In addition to the Sendai mid-term review, the UN Secretary General has delivered Our Common Agenda (2021), a report that highlights the need to do things radically differently in order to avoid existential and catastrophic risks created as a result of human activity. Additionally, The UN Declaration on Future Generations (part of the Our Common Agenda programme) has published an Elements Paper, which explicitly talks about the need to mitigate extreme risks as a priority to protect future generations. NZ’s LTIBs could look much further than the 10-15 years that concerns the present generation.

In addition, the change in emphasis towards greater integration is reflected in the Framework for Global Science co-sponsored by the UNDRR and ISC and supported by the IRDR IPO in Beijing. Overall, the document takes stock of recent developments in disaster risk science and provides a compelling set of directions for research and scientific collaboration for a more holistic and collaborative approach to understanding and managing risks. The framework highlights the rapidly changing nature of risk dynamics, recognizing that a lot of risk work to date has been hazards focused with specific mitigating actions for specific hazards. However, ever increasingly the risks are complex, systemic and interconnected/cascading (See Framework for Global Science Section 5.1, Priority 1).  A concern is that we may remain locked into traditional framings of risks and therefore, overlook interacting, compound, and cascading risks, and fail to effectively manage/plan for catastrophic and existential risks.

These are the arguments for the LTIB signalling that it is part of a future integrated national risk assessment and risk register. This would circumvent the issue of trying to decide what is a ‘security’ risk and what is not, and would help to avoid the problem of institutional silos and would help foster effective communities of practice. To even begin to address the issues of concern, the relevant risks must be included in the analysis.

The LTIB should recommend resources are dedicated to monitoring the four trends identified, but also resources should be committed to acknowledging, analysing and managing the extreme tail risks associated with each trend, in particular those risks that would be unbearable (ie, major nuclear war, powerful unaligned machine intelligence, extreme climate risks (either 6 °C+, or ecological feedforward cascades), and biological engineering of pathogens.

As the most isolated temperate land mass in the world, NZ should be particularly concerned about trade isolation as a severe risk. If any of the extreme scenarios just listed were to manifest, NZ may be the first to be dropped from stressed global shipping or a collapse in air transportation. Such effects could also be brought on by supervolcanic eruption, massive solar flare, large asteroid impact, or a range of other scenarios.

A comprehensive National Risk Assessment would also provide decision-relevant information for rational prioritisation of resources. This is a direction that we believe the National Security LTIB should signal that decision-makers ought to be heading.

Prioritisation decisions are often made within government departments, but shifting resources across government is harder, but more important to do. Areas where NZ is already doing reasonably well probably do not need a lot of extra resourcing (eg, epidemics/pandemics [11], domestic terrorism). But areas where it is perceived we are poorly prepared warrant extra focus, eg, misinformation and impacts of nuclear war/nuclear winter.

The neglectedness of some risks means that investment there will reap low-hanging fruit, while in contrast there are likely to be diminishing returns investing further to mitigate threats NZ is already competent in managing. Government should conduct cost-effectiveness and cost-benefit analyses across national security risks and intervention options (eg, there is scope to significantly advance preliminary work we have done on the cost-benefit of pandemic prevention [12] now that we have Covid-19 experience – this kind of approach can be applied to other threats).

We commend the gestures towards institution building presented in the ten features. However, those writing the LTIB are likely to each exist within one of the national security silos, and each sees what they are familiar with. Other silos exist in the natural hazards domain and yet other hazards and threats may fall between jurisdictional silos. NZ therefore needs a comprehensive National Risk Assessment that integrates natural, security, global catastrophic and existential risks. The LTIB should indicate that this is a goal worthy of pursuing, thereby ensuring there is appropriate responsibility and accountability for foreseeing, assessing, preventing, and mitigating extreme risks that cut across the traditional silos of government, including global risks where indirect harm could be catastrophic for NZ [1].

References

1.         Boyd M, Wilson N. Island refuges for surviving nuclear winter and other abrupt sunlight-reducing catastrophes. Risk Analysis (in press).

2.         Preddey G, Wilkins P, Wilson N, Kjellstrom T, Williamson B: Nuclear Disaster, A Report to the Commission for the Future. Wellington: Government Printer, 1982. https://www.mcguinnessinstitute.org/wp-content/uploads/2016/11/CFTF-March-1982-Future-Contingencies-4-Nuclear-Disaster-FULL.pdf.

3.         Green W. Nuclear war impacts on noncombatant societies: An important research task. Ambio. 1989;18:402-406.

4.         Green W, Cairns T, Wright J. New Zealand After Nuclear War. Wellington: New Zealand Planning Council, 1987. https://www.mcguinnessinstitute.org/wp-content/uploads/2019/11/A-New-Zealand-After-Nuclear-War.pdf.

5.         New Zealand Planning Council. New Zealand after Nuclear War: The Background Papers. New Zealand Planning Council, 1987.

6.         Boyd M, Wilson N. Assumptions, uncertainty, and catastrophic/existential risk: National risk assessments need improved methods and stakeholder engagement. SocArXiv 2022;(5 August). Doi: 10.31235/osf.io/jt28k.

7.         Ord T. The precipice: Existential risk and the future of humanity. Bloomsbury. 2020.

8.         Wilson N, Cassidy M, Boyd M, Mani L, Valler V, Brönnimann S. Impact of the Tambora Volcanic Eruption of 1815 on Islands and Relevance to Future Sunlight-Blocking Catastrophes. Research Square (Preprint) 2022;(10 October). https://www.researchsquare.com/article/rs-2124163/v1.

9.         Hellman M, Cerf V. An existential discussion: What is the probability of nuclear war? Bulletin of the Atomic Scientists. 2021;(18 March). https://thebulletin.org/2021/03/an-existential-discussion-what-is-the-probability-of-nuclear-war/.

10.       Boyd M, Wilson N. Anticipatory Governance for Preventing and Mitigating Catastrophic and Existential Risks. Policy Quarterly. 2021;17(4):20-31.

11.       Boyd M, Baker MG, Nelson C, Wilson N. The 2021 Global Health Security (GHS) Index: Aotearoa New Zealand’s improving capacity to manage biological threats must now be consolidated. The New Zealand Medical Journal (Online). 2022:89-98.

12.       Boyd M, Mansoor O, Baker M, Wilson N. Economic evaluation of border closure for a generic severe pandemic threat using New Zealand Treasury methods. Aust N Z J Public Health. 2018;42:444-446.