The eagerly awaited review of government statistics by Professor Sir Charles Bean, published this morning, is an landmark report that should help the ONS refocus on core activities and do them more professionally. It advocates access to new data sources and the use of state of the art technologies.
The report sets out the problems that the ONS has had in recent years – some elements of which are shocking – and says that a cultural shift is required for change. It is an opportunity for all stakeholders and users to debate ONS work priorities priorities in advance of Bean’s final report in March. That report will tackle governance arrangements. Given the state the stats are in and the task list now facing the ONS, UKSA is going to have to change if it is to survive in its current format.
- Refocus the culture of ONS towards better meeting user needs
- Make the most of existing and new data sources and the technologies for dealing with them
- Become better at understanding and interrogating data
- Address established statistical limitations
- Become more agile in the provision of statistics that properly reflect the changing structure and characteristics of the economy
The report is forthright about new data sources, saying: “It’s nonsensical that different bits of the government don’t speak to each other” and that the UK lags other countries in the use of its own data. It adds: “Using government administrative data could help improve many economic statistics, including providing more accurate early estimates of GDP, improving regional statistics, and providing more insight into financial, trade and labour market flows.” The Report recommends that new legislation is required “to overcome existing barriers and help open up all publically-held microdata to ONS to improve their economic statistics”. This is not new but UKSA seems not to have given this a push. It must now act to help the ONS deliver.
Delivery will need the right people and technology. The report “recommends ONS invest in technology and staff, including a cadre of data scientists. To ensure ONS moves ever closer to the cutting edge, the report recommends that it should constantly be on the lookout for new data sources and techniques, learning from businesses and statisticians everywhere.”
The ONS will need a mind-set too. “A culture shift at the ONS is also key to producing economic statistics for a modern economy. It needs to become an organisation which is more intellectually curious, open and self-critical, as well as better at engaging with its customers.” It said that there was a need for the “capability to quality-assure and sense-check statistics before release”. It says that the “ONS, and other producers of economic statistics, need to move away from focusing largely on the production of statistics and become rather more of a service provider, helping users answer their questions about the economy.” (para 1.14)
Expressing some frustration with the failure to follow through fully the actions of previous reports such as those authored by Allsopp and Atkinson, Bean said that “ONS (should) put in place plans to address these limitations, with a precise timeline for delivery”.
It envisages that ONS should increase its London presence “in order to facilitate stronger engagement with key customers, as well as expanding its engagement with users across the rest of the UK”.
It was deeply reassuring that the report stated that “Reliable economic statistics are an important public good” (para 1.11). There was a concern that, with public spending tight and restrictions set to be placed on freedom of information, data might not be given a high priority.
The bulk of the report’s chapter 2 discusses four long-standing challenges, namely: the construction of GDP; improving the coverage of services (including financial and public services); understanding financial-interconnectedness; and the provision of regional statistics. There is no doubt that these are key themes. Others could have been added – and might be for the final report – including, for example, more coherent labour market data and housing market figures. Some quick wins would be valuable given some issues are going to be very hard to progress.
Governance issues will be the focus of the final report. The National Statistician and his (mostly recently recruited) senior executive staff were given an endorsement in the report but some sort of shake-up of the UKSA governing board seems likely as the report gives the impression it has not always had its eyes on the ball. At one level, if it had been doing its job properly, it could have identified, highlighted and forced a debate on the issues raised in the Bean review so that the Treasury would never have seen it as necessary. At the more detailed level, some of the weaknesses set out in chapter 3 of the report – the squeeze on resources, the weaknesses in the ONS finance function, project management, delivery of recommendations from previous statistics reviews, sluggish repair of the website, data errors not picked up by quality assurance processes, etc – should have been picked up by the UKSA board and dealt with in an open and transparent way.
The link with international bodies is up for grabs. The report says that the ONS has “primarily operated as a ‘factory’, generating prescribed statistics for an increasingly broad set of economic variables from a range of surveys, usually according to methodologies laid down in internationally-agreed accords, and with only modest additional interpretation. The definitions and methodologies laid down in those accords have generally evolved to reflect the changing structure of the UK and other economies, but usually only with a considerable lag.” It says that it is “debatable” whether such a model remains appropriate in the twenty-first century.
The report is a solid foundation for the debate that will now follow.
Terms of reference:
On 10 July 2015, as part of the government’s productivity plan, the Chancellor of the Exchequer, George Osborne, announced an independent review of economic statistics. The Terms of Reference of the Review were to:
- Assess the UK’s future (economic) statistics needs, in particular relating to the challenges of measuring the modern economy (‘Needs’);
- Assess the effectiveness of the Office for National Statistics (ONS) in delivering those statistics, including the extent to which ONS makes use of relevant data and emerging data science techniques (‘Capability’);
- While fully protecting the independence of UK national statistics, consider whether the current governance framework best supports the production of world-class economic statistics (’Governance’).
Bean had been asked to make interim recommendations to the Chancellor and the Cabinet Office Minister in the autumn of 2015, with a final report published by Budget 2016. This Interim Report meets the first of those requests. It focusses on the first and second bullets of the Terms of Reference, leaving governance questions to be addressed in the Final Report. It is also the intention to expand on the particular challenges and opportunities posed by the digital economy in the Final Report.