The school lunches debacle is a classic example of what happens when social policy projects lose sight of their original objectives.

When the school lunches scheme was established in 2018 the intention was to provide a healthy lunch to those students who because of their circumstances might be going without lunch altogether at school – which would be having a negative effect on their learning – or whose lunches were inadequate and non-nutritious. It aimed to build on more localised programmes that had been run by voluntary agencies like KidsCan over the years.

However, concerns that targeting disadvantaged children by providing them with a school lunch might stigmatise them amongst their peers led to the programme being effectively universalised for those schools involved. Even that was a problem – it stratified schools as rich schools and poor schools and made the false assumption that no children from deprived backgrounds went to rich schools, or vice versa.

The change to greater universal provision meant costs grew rapidly – by the end of 2023 the government was spending more than $320 million a year on school lunches, up from $260 million when the scheme was established.

Unsurprisingly, when the current government took office in 2023 and sought to pare back government spending, the school lunch programme was an early target. The government produced a plan to centralise and standardise the school lunches operation which would save $130 million annually. Subsequent events have shown, even after all the criticism and political posturing from schools and disenfranchised former providers is discounted, that this was not quite as simple or straightforward a process as first envisaged. Today, the government is engaged in a major rescue exercise to try to make its school lunches reforms viable and credible.

However, the basic problem is far more fundamental than just the ability of the current national contractor to deliver the government’s revised approach. The seeds of the current crisis were sown when the objectives of the original scheme were broadened in 2019 to include all children in qualifying schools. That inevitably meant that it was likely to become financially unsustainable in the longer term, forcing whichever government was in power to review what was happening.

But sadly, this approach and result is not an isolated instance. The worthy intention of successive governments not to draw undue attention to disadvantaged or at-risk groups that were receiving the benefit of social assistance has led to policy interventions becoming more broad-brush than targeted to where need was greatest.

For example, the world-renowned Dunedin longitudinal study has found that it can identify children and families likely to be at-risk at an early stage. Similarly, Police data shows that there are a small number of families that dominate the criminal offending statistics. The implication of both is that we have now the data to target specific policies and interventions around those families.

But successive governments – except for Sir Bill English’s short-lived government and its focus on a social investment guiding social service provision – have shown no interest in wanting to follow such a data-based approach to social policy.

Indeed, if anything, their approach has been towards greater universalism, and more broadly based policies. The net result has been more and more resources wasted on people who do not really need government support, while the most vulnerable and at at-risk do not get the help they need, because the government cannot afford the burgeoning cost of universalised social services.

While this broad-brush approach remains the focus, it is inevitable that most social policies will fail to significantly address the needs of those at whom they should properly be aimed. And that will give rise to questioning of the way in which the services are being provided, rather than whether they are being aimed at those that need them. This is precisely what is happening in the school lunches debate at present.

If the government was genuinely brave and serious about reforming social policy to ensure it best meets the needs of those at-risk, it would be looking to a fundamental, back to basics, reset along these lines.  However, that will never happen because too many people now gain from the smothering web of universal social assistance, and no government will want to be seen to be taking something away from people, even if they do not need it.

The perverse outcome of all this is that the current system entrenches division and deprivation, rather than resolves it. The big losers in the school lunches mess remain the group no-one seems willing to talk about – the kids coming to school each day, poorly and inadequately fed. There has been much discussion about previous providers who have lost their contracts, school principals finding the transition to a new system difficult, or even some students finding the meals do not meet their tastes. But there has been little obvious consideration of the needs and responses of the genuinely at-risk children for whom a school-lunch is not another “nice to have” provided by the state, but a genuine necessity.

So long as the current “all things to all people” approach to social services provision continues, programmes such as school lunches will keep failing to meet their objectives, however they are structured. And the level of public dissatisfaction at their mounting cost and apparent ineffectiveness will increase.

Yet, the potential is there to make genuine change for the good by using the data now available to develop targeted programmes focusing on meeting genuine need ahead of just making people feel good. Hopefully, one of the lessons emerging from the school lunches debacle is the imperative to concentrate on resolving real, genuine evidence-based need, ahead of any other factors.

But the uncomfortable truth remains that feel good universalism ultimately satisfies nobody – and leaves the basic challenges of inequality and deprivation unresolved.