Things do get better
Why we should all love pensions policy
First: apologies. I’ve been quiet for just over a month, and for no particular reason except that I found I was working too hard and couldn’t quite convert my emerging thinking into coherent writing worthy of your time. I’ve made a list at the end of this piece of the substacks-I-half-wrote-in-June and I’d love you to let me know which you’d most like to read.
Later today I’m doing a talk with the wonderful Will Sandbrook at the NEST Insight conference about how change happens, and how think tanks contribute to the process. Preparing for it has relit my confidence and optimism about what remains possible in our complex, conflicted society, and how those of us committed to change might start to organise ourselves differently to get things done.
This is partly a story about pensions policy. But mostly a story about what the rest of us can learn from what has been handled well, not just once but repeatedly for nearly twenty years. Of course: pensions are not solved in the UK. Problems don’t get solved. Society is not perfectable. But pensions have got better, in substantial and meaningful ways and when most of our politics is spent shouting about failure and polycrisis, we should let ourselves pause to notice that more often.
You might want to note that James Purnell, the Chief of Staff to the incoming Prime Minister, plays an important role in this story. Those of you inclined to optimism might smile quietly. But those of you inclined to hero-worship should not get too excited: the magic belongs to no individual, but to collaboration and imagination.
What went right
In 2002, the government set up a Pensions Commission under Lord Turner to address the chronic problem of under-saving for pensions. It reported in 2006, after extensive consultation and engagement, recommending an increase in the State Pension age and the establishment of automatic enrolment for all employees into workplace pensions to which employers must contribute. The former was controversial with workers, the latter with employers: the deal was a partnership.
Of course, the Pensions Commission has been held up as a policy success for a long time: this note by the Institute for Government summarises the findings of a review they did back in 2010. So celebrated is the Commission in policy circles that the Starmer government revived it last summer to help address the next generation’s pension problems.
The simplest story is one of expertise: bring in some clever outsiders who can navigate the complex politics of controversial, long term change. My interest is in some of the details that get a bit less attention.
First: for ministers - as the IfG reports, it was a lot of hard work, including for ministers. John Hutton, the Secretary of State, and James Purnell, the Pensions Minister, played critical roles. The PM himself devoted 5-6 hours a week during crunch periods to getting the deal done.
Second: while politics has been all over the shop in the last twenty years, on this the politics held firm. David Cameron did wobble a few times during the coalition years as auto-enrolment kicked in and businesses grumbled a bit about the costs. But he never surrendered to the temptation to sabotage long term gain for immediate benefit that was indulged on the topic of social care reform by George Osborne himself in 2009 (“death tax!”), and by Jeremy Corbyn (“dementia tax!”) in 2018. Why?
My hunch is that it was the central deal which helped: matching the stronger pension system with the increased pension age. No side wanted to put their gain at risk by breaking the link.
Third: the investment in narrative building has been extraordinary. The original Commission spent years sharing evidence, constructing dialogue and engaging with the public at scale, and in detail before they finalised their recommendations. But the story didn’t end there. Serious resources have been poured into advertising the scheme - in remarkably clever ways that tell a story of mutual collaboration. When you put money in, your employer does too, and the government. It’s a story of citizenship and participation - a story of give and take, not dependency or entitlement. That narrative building has a public policy objective: to get people to stay in the system. But it also serves a wider agenda of telling people something about what it might mean to be part of a society - something government should do far more of (as I argued here).
Fourth: and this is what interests me most of all - this has been an exercise in institution-building.
NEST is the UK’s largest pension fund by membership; it is the default pension scheme for auto-enrolment. It was set up because it was clear there was a gap in the market that needed to be filled by a mission-driven organisation. One that could act in the market - NEST invests in all the normal things that pension funds invest in - but could serve a set of customers (both pension savers and employers) who were just too small and complicated to be viable for the mainstream pensions sector.
Institution-building is an under-developed skill for the British state. We either try to outsource to the market, or insource to the public sector, under-valuing the possibility of creating something that has the strengths of both. NEST was ambitious not just because of its intent, but its model: the savings trust is owned and controlled by a public entity - with a clear public interest mandate. This institution-building is another thing I’ve argued we should do more of: a starting point, not a last gasp idea for what quality capacity building should look like.
Nest Insight
And that’s where NEST Insight comes in. They are the in-house think tank of the NEST system, and of course they are not the main thing that NEST has achieved in the last 15 years. But:
I think their change model is more interesting than most think tanks.
Their very existence demonstrates the value that public interest corporations can create because of their liminal status - belonging to the nation but not fully a part of government.
On the model: NEST Insight has made the most of its relationship with its parent organisation. Instead of thinking about change, they have played a part in it. They have run large trials and pilots with customers and partners; they’ve used behavioural insights to test, learn and grow what works; and they’ve used those live insights to influence policy upstream and help DWP build wider models for financial inclusion and resilience across the population.
Yes: they are able to do this because they have the privilege of operating inside a giant consumer services business. But NEST’s privilege is a feature of good institution-building, not an accident. Government can’t lobby itself, and when it publishes research findings they often become hugely controversial. Building ‘test learn and grow’ capacity within government is essential but challenging - as Nick Kimber writes beautifully here. When the coalition tried to create a ‘contestable policy fund’ to commission policy work from outsiders, most of the civil service loathed it and tried to sabotage it.
Think tanking is national capacity, but it is weak in the UK. We are chronically underfunded and undercapitalised; in much of Europe state funding goes to think tanks; in the US, vast quantities of philanthropy support policy making. The UK landscape is patchy, dependent on short term funding, commercial research income and (too often) selling of political access rather than policy insight.
So the genius idea of adding think tank capacity to the build-out of NEST should be celebrated. We should do it again. Where else do we need teams of ambitious, data-rich, liberated thinkers? I’d love to know what nationally-funded think tank capacity you’d build, if you had the chance.
Unwritten substacks:
We need a new model of vulnerability. The debate about banning social media for the under 16s highlights the chaotic way we think about human capacity - absolute in most circumstances, absent in others. From Power of Attorney to the Mental Capacity Act, and now these rules for children, we want to draw fixed lines that make life easy for systems. But human capacity and vulnerability are dynamic and complex. What would it take for us to construct alternative legal models appropriate for regulating addictive and habit forming products and services in the light of a deeper model of human frailty?
Mavericks. Steve Hilton, my old deskmate, is running for Governor of California and for a couple of weeks I was bombarded by journalists asking for anecdotes about him. I disagree with Steve on multiple fronts and his temperamentalism got in his way when he was in government. But I’m still not a hater: I think mavericks have a critical role to play in disrupting stuck systems. But how do you get the value of a Steve (or a Dom) without breaking everything?
Blended finance. I’ve been thinking a lot about money recently: as I’ve mentioned before, I’m involved in building a National Pain Institute and a donor collective on End of Life Care. Both are ‘wicked’ problems: requiring co-ordination and collaboration across different parts of the public sector, as well as between individuals and institutions. They are social as well as technical. And they need a combination of public, private and philanthropic capital. The normal ways we put money into these systems don’t solve problems: they pay for the consequences of those problems. Is re-engineering the financial system the key to unlocking change?


Thanks for this. It's good to highlight (and learn from) just what a success NEST has been - from the initial construction of a consensus for change, to the adoption of "nudge" style auto-enrollment, to subsequent institution building. But for all those achievements we seem to have fallen into the trap of thinking "job done". That's led both policy makers and contributors to assume the existing NEST contribution rate will deliver decent future top-up pensions. But will it? No. The rate would need to more than double to match the rate of a good defined benefit pension scheme. So there's a lot of room for improvement, and the clock is ticking...
Really interesting, as ever, Polly. I really liked your point about how to get the politics right and maintain broad consensus. There's something compelling about the idea of a quid pro quo 'deal' like the one on pensions that might be worth applying to other wicked policy questions - such as adult social care - but perhaps also questions like reforming party funding