• The world is facing a pensions crisis by 2050.

• AI-assisted technology could both increase pensions access and improve financial outcomes for savers.

• Pensions stakeholders must foster trust in such technology.

How can we fix the pensions timebomb? Previous World Economic Forum analysis has predicted at least a $400 trillion savings gap by 2050 – and that’s only across the eight most populated or developed savings markets. Meanwhile, the UN predicts that, globally, the population of over-60s will double to 2 billion people between 2020 and 2050. Around 80% of these will be in “less” or “least” developed countries.

Future generations are faced with the increasingly impossible task of funding the needs of an ageing population, as well as their own future. A challenge that is being compounded by structural issues in retirement funding, shifts in employment stability, and a lack of financial inclusion and literacy.

Size of the retirement savings gap ($ trillions, 2015)
Size of the retirement savings gap ($ trillions, 2015)
Image: Mercer; World Economic Forum

The modern pensions landscape

In 2017, a World Economic Forum paper analyzed the multiple aspects of this challenge. It lists the demographic factors above, but also describes several other obstacles, including long-term low growth for investment returns; lack of easy access to pensions; and the shift to direct contribution (DC) savings models.

This third point is crucial because savings rates are typically 10% to 15% lower under DC. They also require more self-directed investment decisions, and therefore higher financial literacy amongst the populace. While financial literacy is slowly improving, there is still a big gap, and this is worse in younger people, and in less developed economies.

On top of this, individuals find it incredibly difficult to get a holistic view of their finances that is relevant to their long-term savings perspective. A big obstacle is the lack of data-sharing and integration between financial products and providers. The challenge is further compounded by the large number of jobs people tend to have over their lifetime – each coming with its own different pension plans.

We believe technology and automation tools that coach positive behaviours and enable easy long-term financial planning are a necessary part of the solution. Indeed, this supports recommendations to policy-makers made in the previous World Economic Forum paper, which included a focus on making savings easy for everyone, supporting financial literacy and standardizing pension data.

Better decisions through automation

In the last decade, innovations at the intersection of technology, regulation and behavioural finance have strengthened DC savings vehicles. In large part, this has been done by making the process of investing more automatic for individuals.

Automatic enrolment sets up employees to contribute to their retirement as a default. Automatic escalation bumps up savings rates slightly each year. Together, these can increase retirement savings readiness by well over 33% and improve savings rates by 56% (see below). Target date funds, used widely in the US, automate the process of investment selection for individuals, leading to more balanced, risk-appropriate choices.

Fintech can improve retirement prospects
Fintech can improve retirement prospects
Image: How America Saves 2020: Insights to Action; Vanguard

And even more can be done through creating a personalized and holistic view of an individual’s finances, combining this with artificial intelligence technology that supports appropriate and responsible day-to-day nudges toward long-term financial thinking.

This “nudge” thinking has been central to many “fintech” business models, particularly in retirement savings. For once, the technology is not the barrier. It enables scalable mass automation, meaning individuals can access low-cost but personalized recommendations – opening up financial coaching to a much wider audience globally.

The integration dilemma

The real challenge to optimizing this benefit is in encouraging interoperability between financial providers. The challenge is not technical, but rather one of competitive concerns within the industry. Who wants to let their customers easily compare their financial products side by side with their competitors?

Lessons can be taken from the experience of the UK’s Open Banking regulations (based on the EU’s Revised Payment Services Directive). This requires banks to release their data in a secure, standardized form so that it can be shared between authorized organizations online. The default method has been for banks to use open Application Programming Interfaces (APIs) that enable customers to share their data across different banks, often aggregating them in new fintech apps or platforms that help with money management.

While the full success of this approach has yet to be determined, it does show that, with regulatory backing, the financial industry can be made to modernize and reshape the competitive nature of their services.

Transparency and trust

For all the potential good technology and automation can bring to retirement savings, none of it will come to pass unless the institutions and stakeholders governing the technology can deliver transparency and foster consumer trust as they deliver and develop these new ideas.

On the transparency front, technology must move fast and be packaged up as an attractive and rewarding experience – but also remain explainable. Users need to understand financial services clearly enough to trust them. Automation requires consumers to disclose personal information – and to be comfortable with an algorithm providing advice.

Individuals, regulators and experts will also need to be able to audit, inspect and monitor the “black box” that provides advice or direction to individuals, ensuring it is appropriately stress-tested and free of bias.

Stakeholders from government, private enterprise and non-governmental organizations ultimately hold responsibility when retirement savings fall short. Technology and automation, moderated by empathetic human hands, hold the potential to guide more and more workers to grow into savers and make better decisions in each stage of their life journey.

Artificial Intelligence

What is the World Economic Forum doing about AI?

In 2019, the World Economic Forum’s Centre for the Fourth Industrial Revolution convened an informal multi-stakeholder group of leaders, known as the Global AI Council (GAIC) a keen interest in creating positive futures with advanced AI systems.

One of the goals of the Council is to provide strategic guidance to the global community on the priorities for AI governance and cooperation as well as the policy implications linked to advances in AI.

The project is taking place over several months and brings together a diverse group of individuals that includes science-fiction authors, economists, policymakers, and AI experts.

The council aims to open up the possibilities for its Positive AI Economic Futures using the creativity and expertise of these participants as well as opening up the process to a much wider range of contributors.

It is also in the process of initiating a second thread of the project, running in parallel with the workshops: a movie competition in partnership with the XPRIZE Foundation. Participants will create short movies showcasing their ideas for a future economy in a concrete form that speaks to individual aspirations and fears.

Together, under the aegis of the World Economic Forum, behavioural scientists, financiers, pension advisors, technologists, ethicists, consumer groups and regulators may help create a framework that engenders trust and transparency, which will enable technology to improve retirement outcomes for many individuals and their communities.