We are starting a new decade with a new government and fresh debate about what skills the civil service needs. One angle on this debate is about how best to answer the age-old question of “What makes effective policy?”

Various themes in answer to this have risen and fallen in popularity over time: “evidence based policy” and the emergence of the What Works Centres; an emphasis on “nudge” theory and behavioural insights; delivery units and implementation units; and the ebb and flow of an emphasis on project and programme management disciplines, delivery chain analyses, and devolution / localism vs centralised control.

Whilst individual disciplines and specialisms have developed beyond recognition in the academic sphere, many of the key policy questions remain the same, and just as elusive as ever. Where should public funding be invested to make the biggest possible difference? How can we be confident it will have the effect we expect? How will we ever be able to prove it?

In the late 1990s I was involved in the Treasury’s cross-cutting review of services for young children. A review of the evidence at that point illustrated through thematic papers and a great deal of consultation that services were often patchy and incomplete, and failed to provide a coherent set of support particularly for the most vulnerable families. The “best” evidence available in terms of rigorous analysis at that stage came from the USA’s evaluations of the HeadStart programme, and some great analysis of UK cohort studies which showed huge potential benefits for society of getting things right in the early years. That review produced Sure Start Children’s Centres – but what got less attention was its recommendations for better joining up of services across local systems.

The process of writing recommendations from that review benefitted from a range of external academic expertise, input from an expert advisory group, and a series of commissioned thematic papers by experts. But I don’t think we ever properly and publicly acknowledged the value of the informal professional input from one extremely experienced children’s services manager in the voluntary sector who spent many hours on the phone helping me to make sure that the recommendations and subsequent guidance made sense to real people in the real world – triangulating the different professional and academic perspectives to draw out the most important points and principles which all our work suggested would together make the biggest difference to helping very young children and their families. It wasn’t until many years later that I realised how lucky I had been to have the patient and careful input from that lady at that point. Thank you Ann!

But the wider lesson here is that some of the most successful policies – which have, ultimately, to be understood and be backed by those on the front line – are those which are co-developed by policy makers with people working in the field, service providers and commissioners, academics and other experts, and ideally also with people using public services or who are themselves subject to the policy. And this is particularly important when the aim of a policy is to help groups affected by a range of different issues. Individuals and families who are struggling with unemployment, debt, mental or physical health problems and alcohol misuse are likely to be subject to a wide range of policies and have to work with a huge range of different agencies. Whilst “joining up” those services is clearly a huge part of the solution, the design of policies themselves need to be tested early on for how they interact with others. It is only relatively recently that research has shown that well-evidenced parenting programmes don’t work for families struggling with domestic abuse. But designing a policy solution to that issue needs to involve children’s social work, child development experts, an understanding of therapy and much wider expertise; it cannot be solved by one discipline alone.

So whilst I can see the merit of widening the range of disciplines and skills of people in the civil service, my plea would be that significant policy change should be thoroughly tested through multidisciplinary and multi-sectoral teams. Whilst we won’t always agree, the policy will be stronger as a result of the debate.

Sally Burlington is Head of Policy (People) at the Local Government Association. Previously she worked as a senior civil servant in a range of Government Departments specialising in social policy issues, after starting her career as a Treasury economist.