The prime minister needs to clarify whether he still believes in ministerial accountability

In the UK’s system of government, the impartial, permanent civil service advises democratically elected ministers, who in turn make the final decisions on all policy matters. This division means that, as set out in the ministerial code, it is ministers who are accountable to parliament for the actions of the departments they oversee. Ministers can understandably feel frustrated about this arrangement, where they can end up carrying the can for what seem to have been officials’ mistakes. This was the case when Amber Rudd resigned over the Home Office’s handling of the Windrush scandal – her officials briefed her poorly, but it was the minister who lost her job.

This government appears to be taking a new approach to accountability. In recent months, particularly during the coronavirus pandemic, it has seemed content to blame officials for things that have gone wrong, despite the fact that officials cannot present their side of the story. This approach raises questions about whether the prime minister truly believes ministers should be accountable for what happens in their name.

Ministerial decisions played a part in the exam fiasco and Public Health England failures

Following the recent furore over A-level results, two senior officials have resigned: Sally Collier, chief executive at Ofqual, and Jonathan Slater, permanent secretary at the Department for Education (DfE). The latter’s resignation came after the prime minister ‘concluded that there is a need for fresh official leadership at the Department for Education’ ^[1], an intervention implying a judgment of poor performance which has not been made in such a public way before.

It is impossible at this stage to know exactly what happened inside Ofqual and DfE that led to the debacle over standardised exam results. The education select committee is due to hear from both the secretary of state, Gavin Williamson, and Roger Taylor, chair of Ofqual, in September. The committee must find out what advice was given and when, and how ministers responded. A minister is completely entitled to override warnings and advice from their officials and from outside organisations. But if their decision to do so leads to a negative outcome, parliamentarians and the public are entitled to ask why they took that decision. Requiring a permanent secretary to step down does not remove the responsibility from ministers.

The planned winding-down of Public Health England (PHE) and the creation of the new National Institute for Health Protection (NIHP) also reflect the government view that it was official structures, rather than the minister in question, that were responsible for mistakes made in the early stages of the pandemic. While mistakes may have been made before and during the crisis – as was inevitable in responding to a new disease – PHE was working to the direct instructions of ministers all along. And ministers will still be responsible for what the NIHP does. The official inquiry into the government’s handling of Covid-19 will need to uncover what instructions were given by whom, and how they were implemented, so that lessons can be learned.

The government must be open about its mistakes

There are occasions, as we have argued, when a secretary of state must take responsibility for the failures of their department. However, taking responsibility is not the same as achieving effective accountability. As previous Institute work has argued, when it works well, accountability is not just about apportioning blame but promoting improvement in how government works, promoting citizens’ confidence in their government and increasing the wider legitimacy of the public sector.

If our system of accountability is to work properly, the government must be open about who was responsible for what and be willing to learn from mistakes. And if Boris Johnson still stands by the ministerial code, which is issued in his name, then he should stop allowing officials to take the blame for everything that goes wrong.