Evidence from Engagement with ONS

OSR has engaged with ONS to test and explore issues raised from previous assessments of ONS economic statistics and by stakeholders. We have reviewed ONS-published strategies, reports and other documents (such as survey quality and methodology information statements). Additionally, we have held focus groups and bilateral discussions with around 40 ONS staff, largely based in ONS’s economic statistics divisions.

Invitations to participate in engagement with OSR were issued to individuals following agreement between OSR and ONS on the roles whose post-holders would be best placed to make an informed contribution to the understanding of issues raised by stakeholders. Since our review has focused on data sources, our engagement with ONS staff has so far reflected this focus, though the roundtable discussions were not narrowly constrained to this area.

The views of ONS staff were summarised and circulated to participants in focus groups and bilaterial meetings, with issues where there was a strong consensus identified. Participants were assured that their views were being sought on a non-attributable basis.

Many of the issues summarised so far in this report have been acknowledged by ONS staff in an open and non-defensive way. Concerns about the effectiveness of processes for allocating resources and prioritisation were recognised, as was the slow progress made in addressing the Bean recommendations on measuring the modern economy.

Additional issues raised by ONS staff included the following:

• Despite the existence of a risk-based strategy for quality, many staff thought there has been an erosion of quality in some areas (with, for example, less frequent survey reviews and sample re-allocations than in the past and the lack of a systematic, prioritised approach). There were also calls for additional strategic focus on the quality of data inputs in addition to work done to expand the coverage of outputs and to improve methodology.
• A significant number of staff expressed concerns about a claimed failure to recognise and reward expertise adequately, and the associated loss of expertise to the organisation. Such concerns were even more pronounced among staff at the G6/G7 level than the Senior Civil Service (SCS) level but present in both groups.
• Obstacles created by inflexible and unresponsive IT systems were widely noted. Staff thought that these problems were exacerbated in part by ONS’s policy choice to prohibit updates to the legacy systems on which some key business surveys depend.
• Some staff referred to an organisational culture which, despite recent encouragement from senior management, does not always promote or reward challenge and curiosity. Early warnings were not always appreciated or encouraged. As we finalised this interim report, ONS promptly and transparently reported errors in Trade in Goods, Trade in Services and Producer Price statistics. We will consider whether there are any systemic lessons to be learnt from these errors in our follow-up work.
• Several staff noted that, while the availability of economic expertise has increased, there may have been a failure in some areas to fully exploit economic expertise, and the expertise of other specialists, to increase the level of organisational curiosity, analytic oversight and challenge. It was also recognised, however, that in some areas there is extensive engagement of economists, including from other parts of government. Staff also highlighted examples of good practice, such as the running of regular “curiosity” sessions to provide challenge.

Some staff thought that structural issues contributed to the challenges that ONS faces. It was argued that the separation of statistical production, survey data collection and administrative data sourcing into separate directorates might inhibit a joined-up and strategic approach.

Staff also referred to the pressures on them that have resulted from the level of resources available in some areas, including, for example, unfilled vacancies. Some thought that the persistence of unfilled vacancies resulted, at least in part, from financial constraints, either on ONS as a whole or on those parts of ONS responsible for economic statistics. Several members of staff argued that the balance of resources between economic statistics and other areas does not reflect ONS’s core obligation to produce quality economic statistics.

In respect of the fall in response rates, which particularly affects household surveys, ONS staff pointed to the steps being put in place to remedy the situation. These steps include increasing the number of field interviewers and sample sizes; creating communication campaigns; increasing incentives; and investigating pay and benefits options to improve interviewer retention. But staff thought that increased funding in this area may be needed to drive further improvements.