On first reading, the latest Carnegie report on wellbeing does not offer much cause for optimism. It shows that wellbeing in England, measured by GDWe, was falling before the pandemic, even as GDP was rising. It focuses in on democratic wellbeing. And here it shows widespread scepticism and mistrust in democratic institutions. It concludes in starkly worrying language that there is a “collapse in wellbeing”.

It also has some strong recommendations for the Office for National Statistics, related to updating the comprehensive wheel of wellbeing more regularly. These are important points and, like all user feedback, deserve to be looked at seriously by ONS. We will take forward discussion with ONS, as their regulator, on how they respond to user perspectives like this.

But putting the worrying findings to one side, and the ONS recommendations, are there reasons to be positive in this report?

The answer, for me, is yes. I see two and a half reasons for positivity.

The main one is that, buried in the report, is a striking finding about official statistics. The report shows a lack of trust in government and MPs, and yet there appears to be a degree of trust in official statistics – 61% say they trust the information they see or hear from official statistics. On the face of it, this is remarkable: official statistics emerge from the Government’s Departments and agencies, the very organs of a state that appears to incite so much mistrust. How then is this apparently magical separation, between mistrusted institutions and trusted statistics, possible?

One answer may be that people respect and understand the independence of the Office for National Statistics. Many of the key figures – on the economy, and labour markets, and the pandemic – are collected and published by ONS. But not all are – for example many of the figures on the pandemic are produced by health departments and agencies in the UK’s four governments. It may also be that the questions are different – the question on Government is about a generalised level of trust; whereas the question about official statistics relates to trust information, which is more specific and focused.

But we at OSR think there’s another, more profound reason. It’s that Government statistics should be produced in line with our Code of Practice, which focuses on three pillars – Trustworthiness, Quality and Value. And the Trustworthiness pillar takes as its explicit aim how Government can convey that its statistics are produced, free from vested interest, by statistical professionals in Government. And it embeds this Trustworthiness in clear, credible commitments – for example to pre-announce the publication of statistics and to stick to those commitments; and to be clear in how people’s personal data are safeguarded and used.

We think this is why trust in statistics can be high even as other measures of trust crumble. It is consistent with other evidence we’ve seen, including as part of our public good research programme – and we’ll get a further update later this year.

And I think this is more than just good news for stats. It can also be good news for Government more widely: focus not on pleading for trust, but on the credible commitments, the clear testable evidence, of trustworthiness, and you may start to shift the dial.

The second piece of good news is in the public’s interest in and awareness of public engagement and deliberation. In OSR we’re big on this. And it’s fantastic to see that this is something that members of the public are keen on, understand, and are knowledgeable about. I’m impressed too with their knowledge – 1% of the population even know what a charette is!

This keenness for engagement, and the fact that there are so many tools (charettes included) for securing engagement, is a platform on which to build.

And the half bit of good news? Well, we love user responses to statistics, even if critical. They are the fuel of improvement. And you may wonder – why is this only a half? It’s because user responses are great, but only half way to the improvements. The producers of statistics have got to take on board and act on them.

So, there you have it. Even in an apparently dark hour for trust, there are reasons to be cheerful.

Ed Humpherson

Head of the Office for Statistics Regulation

Photo by Ruthson Zimmerman on Unsplash