Three Big Ideas #12
Fixing flawed science, learning from catastrophes, and building a better Whitehall
Welcome to our weekly Three Big Ideas roundup, in which we serve up a curated selection of ideas (and our takes on them) in entrepreneurship, innovation, science and technology, handpicked by the team – and today, with a guest contribution!
🧪 Eamonn Ives, Research Director
We owe a lot of our safety to the fruits of scientific research, but conducting that research isn’t always safe. Laboratories contain equipment which if used incorrectly or accidents occur can pose considerable risks. Relatedly, suppose some apparatus was even slightly faulty – whatever it was used to produce may end up not quite being what its operator intended.
In a recent commentary article, co-authors Jennifer Byrne and Adrian Barnett describe a more intangible, but no less real, hazard that scientists must contend with – an ‘unsafe’ research literature. They note that the universe of academic publications which informs researchers’ – and by extension all of our – understanding of the world is not always perfect. At best, bad research might see resources squandered as academics forlornly venture down rabbit holes. At worst, it has potentially serious consequences – for instance if it leads to drugs or medical practices that harm patients being approved.
Yet while a scientist can relatively easily flag broken lab equipment to their fellow colleagues, doing the same for ‘broken’ research is altogether more difficult. As Stuart Ritchie (who directed me to this commentary via his always excellent Science Fictions) has highlighted time and again, the process for retracting poor research is mind-meltingly bad. This keeps useless or actively dangerous findings alive for far longer than they ever should be.
Not all innovation can be traced back to the lab, but plenty can. If we’re to have a more innovative future, getting better at fixing each aspect of the research process is a must – and that includes how we weed out unsafe papers.
🚢 Philip Salter, Founder
At 1.28am, on 26 March this year, the container ship Dali struck one of the main piers of the Francis Scott Key Bridge across the Patapsco River in the Baltimore metropolitan area, causing it to collapse. Six people were killed in the catastrophe, and the wider economic impact of the closure of the waterway has been estimated at $15 million per day. Understandably, the public wanted the bridge rebuilt quickly.
As Aidan Mackenzie writes for The New Atlantic the government shifted gears, with planners announcing a relatively aggressive four-year timeline and an expedited regulatory process. Disaster responses offer a clear contrast to the way things are normally done, argues Mackenzie. He also cites the example of a bridge on I-95 in Northeast Philadelphia collapsing after a tanker truck caught fire in an underpass, halting traffic along the entire corridor. The highway was reopened to traffic just twelve days later, months ahead of most predictions. “Pennsylvania Governor Josh Shapiro used his disaster authority, suspending any law or regulation that would impede swift recovery. Construction crews, working around the clock, engineered and built a temporary road in less than a week.”
The gap between how states act in normal times and a crisis suggests that there is potential to considerably increase state efficiency. In the UK, we saw this most acutely in our response to COVID-19, where the Vaccine Taskforce expedited the development, production, and distribution of COVID-19 vaccines, upturning conventional wisdom of the time.
Not every infrastructure or public health challenge is a disaster of the sort cited above, but what these examples show is that we often already know what’s slowing us down. While calls to cut unnecessary regulations can feel like a tired trope, it’s a trope because it’s true, and it’s tired because it’s long overdue.
⏩ Patrick King, Senior Researcher, Reform think tank
When Whitehall succeeds, so too does the country – with benefits felt across public services but also by charities and the private sector. Take regulatory reform, for example, which through the announcement of a new ‘Regulatory Innovation Office’, the new Government has spotlighted as a key lever for innovation and economic growth. Or the structure of the tax system, which has well-rehearsed consequences for R&D, business confidence and investment, and the UK’s competitiveness on the global stage. These challenges are acutely felt by start-ups: often operating at the frontier of the emerging industries that will determine our economic future.
The talent Whitehall has access to is fundamental to its success. Often this involves bringing in people with expertise from outside of Government. But it’s also essential to recruit and develop the best officials – who can work seamlessly across policy and delivery, and effectively unlock barriers to growth and other policy goals.
With billions of pounds of economic output at stake, the pay of decision-makers is one of the last places we should be penny-pinching. Talent comes at a price, and government should be prepared to pay for it.
In a report we’ve published today, Joe Hill and I set out a new model for the Civil Service Fast Stream, which aims to get top graduates into government roles. Despite being badged as a way to recruit elite talent, it has lost this clear focus – both in terms of who it accepts, the training and development it offers, and in its pay and status compared to the leadership schemes of many companies in the private sector.
For entrepreneurialism and the economy to thrive, we need a Fast Stream – and Civil Service – that builds the knowledge and hard-edged skills required to develop effective policy. Of course, there’s no silver bullet; separate to this paper, Reform has outlined how Whitehall can improve the value for money of its spending, better implement AI in public services, and become more mission-led. But an overhaul of the Fast Stream – with more competitive pay and a revitalised development offer – will pay big, long-term dividends.
 | A guest post by
|