Created by Rishabh Srivastava, Founder of Loki.ai
This summary was largely done for my own note-taking, sharing it just in case it adds more value to other people.
I have no affiliation with either the organisers or the presenters. This is a summary largely taken for my own reference, and may contain errors :)
Context
Source URL:
Why is it important: The world is fairly complex, and is getting much more complex
Keywords
Complexity, Governance
Summary
...
Highlights
A. Dealing with complexity in public policy responses
By Aaron Maniam. Deputy Secretary, Industry & Information, Ministry of Communications and Information of Singapore (MCI)
Complexity = situations where we are not clear about cause and effect relationships. 3 Types of complexity out there relevant for public policy issues
- Social complexity: My definition of a problem and its solutions may be very different from that of another person. There's difficulty in defining problems, causes, and effects and what is beneficial. Example: environmental issues
- Dynamic complexity: The cause may be felt today, but the full effects of that cause are only played out later in time. Example: demographic issues. ageing and birth control. Singapore's stop at 2 policy is an example of this
- Generative complexity: Refers to the emergence phenomenon. Seemingly similar events can result in very different outcomes because of very minor differences in them. Example is Covid-19. Very similar policies have led to very different outcomes in different places
We can't "manage" complexity. Rather, we must recognize that the world is turbulent, we need to be agile in it, and the best way to be agile is to constantly experiment and iterate within our public policy.
Rather than enact a certain set of policies and expect them to play out in a stable and predictable way, we need to have an experimental mindset. We need to try small scale tests and prototypes. Beta versions of policies. Once we have implemented these small scale tests, then we can use these experiments to get data about the actual implementation of these ideas. Once we have looked at that data, we can refine our ideas, and eventually get to better ideas
Experimentation is not natural in a public policy setting. Public policy people enjoy having things like blueprints, 5-year plans, working committees that are guided by strategic committees that are guided by planning committees etc. These work for straightforward problems, but when we're faced with social, or dynamic, or generative complexity, we need to adapt.
We need to be okay with failure. If we don't allow ourselves to fail, we don't allow ourselves to learn from those failures. The key is to fail in the most meaningful ways. That we fail when there's actual complexity, and not because we have been complacent about the situations that we have been in.
Fail small, fail fast, and know why you failed.
When faced with complexity, we have to be okay with unordered responses rather than mechanistic or rigid responses.
B. Setting up innovation units
By Marcel ‘Otto’ Yon, Founder, Bundeswehr Cyber Innovation Hub, Germany
2 things that would help deal with complexity – 1) a shift to an agile culture and 2) much more diversity in teams and specially leadership teams. 3 activities that they tried:
- Foster a sharing economy mindset. Share resources, best practices, playbooks across departments
- Build a "lean startup" mentality in the military space where you're okay to try and fail. Also, talk as frequently to people at the bottom of the pyramid and learn from them
- Empower bottom-up innovation. Let platoons innovate in a controlled manner and share their learnings with the high command
Communication is always an underestimated challenge
Key lessons learnt
- Be clear on whether the mission of the innovation unit is to operate within the box or out of the box. Ideally, the innovation ideas should not just be dependent on the system
- Don't hand over of projects to BAU teams too quickly. If the innovation teams tries out an disruptive MVP and hands it over the "line managers" as soon as they can, the project typically fails. They should incubate the project and innovate on the business model until they prove that it can really work
- There are no synergies with the core organization. Innovation unit should not rely on central functions like HR, Finance etc from the main organization. It will simply end up adapting their ways of thinking. The innovation unit must be completely independent
The right way to think about an innovation unit is to think of it as a competitor. If it doesn't act as a competitor to the existing company, it means that something is wrong. It should create friction and constructive conflict. Public sector has to wake up to this. Don't try to standardize everything too early. Have the freedom to allow for competition
C. Types of complexity that governments face
By Piret Tonurist, Innovation Lead, OECD
Governments have to work at different speeds at the same time. Dealing with the problem at hand (reactive/adaptive) and fixing underlying issues (proactive/anticipatory)
Reactive Government
- Position of "wait and see" or called forward when "hazards" (moral,ethical, physical) materialize
- "End-of-pipe" interventions. Often fail to anticipate or address long-term systemic implications
Proactive Government
- Government as a "technology maker"
- Government anticipating various failures and actively exploring and sharping them in practice
Government needs different kinds of innovations at the same time
Anticipation: the creation of knowledge about the future, drawn from existing contextual factors, underlying values, and worldviews + assumptions
Anticipatory innovation: Acting upon knowledge of the future by creating something new that has the potential to impact public values
Anticipatory Innovation Governance: Structure and mechanisms in place that allow and promote anticipatory innovation to occur alongside other forms of innovation
Governments need to experiment, but they also need to transform systems in a way that works. There has to be a path from experimentation to scaling to transformation
How do you scale once an experiment shows promise?
The task is to manage innovation portfolios under complexity. Not just individual projects
D. Panel Discussion
If you were building governments today from scratch
First, deal with the past dependencies that have gone one before. Then, figure out how to build bridges to where you want to go. Don't operate in a vacuum
How can we strengthen the public sector’s ability to understand and respond to complexity? What new skills and competencies should we inculcate
Training is super important. Both skills and mindsets. Planning for potential future scenarios (not predict the future) requires imagination, interdisciplinary thinking. Design thinking is a very important skill, as it focuses on the users (citizens) and what they need instead of what government can supply right now. Anthroplogy is going to be key. We already have great skills in econometrics and data, but ethnographers and anthropologists who can extract stories from data are equally important
Additionally,
- Behavioural Scientists: nudges, choice architectures
- Data Science and Analytics: they can complement the work of anthropologists and designers
Complementarities have to come in in individuals teams
How to build an organization that can deal with rapid tech changes?
The military by itself is very agile (troops in the field). Something just breaks in the military bureaucracy, though. Every agrees that the military bureaucracy should change, but it just doesn't happen
The fix has to be in the culture. Process will only take you so far. Learn from others, particularly from those outside of the military
Fix manipulative behaviour. There is a lot of co-dependency on really old and entrenched relationships that just don't work
How do we simplify the innovation process, and ensure that the rebels and innovation units can work with those who want stability?
Diffusion of innovation. Innovation happens amongst the 2.5% of the population. Then to early adopters (13.5%), then early majority (34%), then late majority (34%), then "laggards" (16%)
Good way to understand any large complex system. All of the system is important. Laggards are not morally deficient. Sometimes, they lag for a reason. Because you want to be very sure that something is good before you switch to it in critical scenarios
Which bits of the government are going to totally change in the aftermath of Covid-19?
Otto: EVERYTHING. Government is super dysfunctional. Silver bullet is teams. Specially senior leadership teams. Take people with an entrepreneurial mindset and insert them in these teams
Aaron: a) How we conduct meetings. if we could be more efficient, so much more good could be done. Virtual meetings may be as good. b) Understand the important of possible futures. When we say something can "never" happen, we are just speaking from our deep biases – Brexit and election of certain political leaders has proved that. c) As we move to be high-tech, we don't have to stop being high-touch. We have to be relational and empathetic in the process of using technology
Piret: Change is going to come whether governments want it or not, because technology is changing everything. We want to get rid of unnecessary hierarchy and unnecessary work. Use more common sense. You often don't need Big Data to tell you what the person on the street could've told you in 5 seconds. Savings in this means that government could be redeployed into much more efficient things