Paul Marshall: Six ways to fix our failing political system

This article was originally published on UnHerd on 2 September 2020.

The recent A-Levels fiasco, in which thousands of school leavers were denied a place at their chosen university, was blamed on a “mutant algorithm” by the Prime Minister — a freak of nature accidentally released to roam the country’s classrooms, threatening to ruin everyone’s exams results until stopped in its tracks.

Compelling though the imagery is, we all know that the A-Levels shambles can’t be blamed on rogue formulae alone. The problems go so much deeper than that. I saw some of those deficiencies first hand when I served as Lead Non-Executive Director at the Department for Education between 2013 and 2016. Indeed, I participated in the appointment of the now-departed Permanent Secretary (of which, more below).

It’s clear that if the Government wants to usher in a brighter post-Brexit, post-Covid future it has to fix its own machinery — and this is the very challenge to which the independent Commission for Smart Government is devoted. Launched last month, it will spend the next year exploring how the UK can be better governed. As the recent grades debacle illustrated, there are many questions to which we need urgent answers.

First, where should accountability lie within Whitehall? The system is clear that financial accountability to Parliament lies with the Accounting Officer, usually the Permanent Secretary. Other types of accountability, however, are less transparent. Historically, at least, there has been a tendency for ministers to fall on their sword. This is in keeping with the “myth of ministerial accountability”, which is applied even when the fault is an operational one. The principle must surely remain that those we elect are ultimately accountable — but is it not possible to make a distinction between accountability for operational delivery (the responsibility of the civil service) and accountability for political strategy (the responsibility of ministers)?

Second, why do we so rarely appoint qualified external candidates as Permanent Secretaries? Back in 2016 I preferred two outside candidates for the DfE job. Ironically, one of those rejected was the very person who warned the Department of the risks arising from re-calibrating the exam results, and the other was head of the UK’s largest private exams assessment organisation. Their experience and skills might have proved invaluable.

Third, why do we call them “Permanent” Secretaries? The job title originated to underline their neutrality and distinguish them from Secretaries of State who come and go with each administration — usually too often. Permanent Secretaries are now appointed for five-year terms, yet the term “permanent” implies their position is secure with little accountability for failure, a title unseen in any other walk of life.

In recent months, some mandarins have begun to look less permanent than others, a small dawning of accountability, perhaps. Some commentators have greeted these unusual departures with shock, as though the removal of those who fail — something we expect in any other sector — was a problem. In fact, if anything needs to be made permanent it is proper accountability.

Fourth, what skills should we expect ministers to have? Of course, it is primarily their political talents that propel them to high office and equip them for the job. But what should they actually know about their subject? Should their position not require at least some domain expertise — a rare occurrence? Indeed, would not some management experience help? We appoint most ministers from the increasingly shallow gene pool of the House of Commons, when we could appoint more from the House of Lords (seeking out proven external talent and ennobling the new recruits). Yet this route has not always been successful. So how could we make the recruitment of talented ministers work better?

Fifth, Ofqual is, in Whitehall-speak, an “arm’s-length body” (ALB), of which there are anything between 100 and 900, depending on how you define them. According to the most recent National Audit Office report, the ALB sector “remains confused and incoherent. There is no single list of ALBs across government nor a common understanding of when ALBs should be used”. Furthermore, there’s an estimated £110 billion of spending in ALBs which is exempt from spending controls — an eye-watering sum.

The NAO also says that there is “no collective understanding of what type of oversight is appropriate for different types of ALBs”. This was certainly my experience as a Non-Executive Director at DfE. Ministers were frustrated by the independence of ALBs like Ofqual and Ofsted and unclear about the extent to which they could intervene and interact. Ironically, ALBs were created to ensure that those with operational responsibility would be discrete, expert and accountable. In practice they result in no one knowing where accountability lies at all. When are ALBs justified and how should accountability mechanisms work?

Sixth, and finally, how can we improve the way Whitehall handles data? While the infamous algorithm has become a running joke, the fact is that we need to be able to make use of more data, and do so with confidence and common sense. Anybody versed in data knows that the outputs are only as good as the inputs, and that any algorithmic outputs need to be subjected to the scrutiny of common sense.

To paraphrase Gary Kasparov after he had been defeated by Deep Blue, “a machine beats a man but a man plus a machine beats a machine”. Yet how many in Whitehall, including politicians, have the numeracy to be able to interrogate assumptions, outputs and inputs? How do we address the numeracy deficit in our leaders? Increased use of data and technology in public services will require new and improved ways to upskill ministers and officials.

The Chief Secretary to the Treasury, Steve Barclay, has thankfully provided new impetus to improve IT capabilities within Whitehall, but we start from a very weak position. There are multiple legacy systems which don’t speak to each other and only 5% of government IT is in the cloud — an extraordinarily low number compared to the private sector or many other governments.

We are in a new era dominated by the knowledge economy, yet many of the practices and structures in Whitehall still hark back to the 19th century. Changes to the recruitment and training of civil servants and ministers alike must ensure that they are competent and equipped to deal with the major challenges that our country now faces.

Smart government has nothing to do with political ideology. Around the world executives of both Left and Right succeed or fail — but to succeed requires an executive that can deliver, offering accountability, incentives, competence and creativity. There is a growing appetite in this country for change — and the new Cabinet Secretary has a big job in leading a long-overdue Whitehall shake-up.