About this article
International review of best practice in the publication of productivity statistics
As part of the UK productivity: analytical release, we are publishing a report on international best practice in the production of productivity statistics. This report, commissioned by the Office for National Statistics (ONS) from the economic consultants, London Economics and DIW Econ, examines the productivity measures that are published by other national and international statistical organisations (NISOs) around the world and includes information on the range, timeliness, detail and frequency of the productivity data they produce. The report summarises the productivity publications of 21 NISOs and includes information on their detailed labour-, capital- and multi-factor productivity estimates.
This report finds that the UK has strengths in the measurement of labour productivity, but there are some areas where we can improve further to be best in class; that we are among the small number of NISOs to have published estimates of capital productivity; and that the work on multi-factor productivity ONS has been under-taking will make us amongst the best in the world for this statistic.
The primary aim of this report was to provide ONS with evidence on international best practice in the production and publication of productivity statistics. This was to identify areas in which ONS is performing comparatively well and to highlight practice from which we could learn. In particular, as we approach the conclusion of the two-year development plan that we set out in July 2016, this report provides important input for ONS’s future plans.
Specifically, this output gives us a sense of where the global frontier of productivity measurement lies and what is required to raise our current practice to that of the world-leaders. The results of this report and the implications for our longer-term development work will both be the subject of discussions at our user group event on 16 March 2018 in London. If you are interested in attending this event, please contact Productivity@ons.gov.uk.
The report finds that against this international yard-stick, ONS’s performance varies by measure. In our publication of labour productivity, our estimates are:
mid-pack in terms of their production speed – ONS’s flash estimate of productivity (published 45 days after the end of the quarter to which they relate) is faster than equivalent estimates in Australia (60 days), Canada (67 days) and Sweden (60 days), but slower than in the US (35 days) and France (31 days)
mid-pack in terms of their industrial granularity – ONS currently produces estimates for combinations of 50 individual industries and aggregations thereof; this compares favourably with Australia (16 industries), Italy (35) and the Netherlands (38), but is considerably less than in the US (380 industries) and Canada (322)
the UK is close to the global frontier in terms of regional labour productivity – ONS is among only six NISOs included in the survey to produce estimates of labour productivity for different regions
the UK is among only three countries that produce industry-by-region estimates of labour productivity
The report also highlights that there is room for improvement in the multi-factor productivity (MFP) estimates that ONS currently produces. At present, we publish multi-factor productivity on an annual basis – around 460 days after the end of the period to which the data relate and among the slowest of the NISOs recorded here. ONS’s current MFP release is also among the most aggregated on an industrial scale of the organisations surveyed for this report: offering estimates of MFP for 10 industries, as compared with 35 in Italy, 39 in Canada and 145 in the US.
The report also notes that ONS is among the small number of NISOs to have published estimates of capital productivity – which estimates the value of output generated per unit of capital input used in production.
The report consequently provides some support for the analytical and development work that ONS has been undertaking over the past two years. In April 2017, we published the first estimates of quarterly regional labour input consistent with the UK’s headline labour productivity metrics. In July 2017, we published experimental labour productivity data at a more detailed level of industrial granularity for the 2009 to 2017 period, as well as the first industry-by-region labour productivity metrics for the UK consistent with headline data.
In October 2017, we published updated estimates of the UK’s labour productivity performance relative to the other G7 economies, as well as the first data on international comparisons of labour productivity by industry. This release extends this run of developments – including the first, experimental real regional labour productivity measures and an extended time series for our more detailed industry level estimates.
The weaknesses in the area of MFP to which the report alludes also provide further justification for the development programme that we have been undertaking and which will come to fruition in the coming months. In July and October 2016, we published estimates of quality-adjusted labour input (QALI) – an important ingredient for MFP – using a new, experimental methodology, which will increase the industrial granularity of QALI to close to 60 individual industries on a quarterly basis. Alongside this release, we have published the first experimental volume index of capital services (VICS) results on a quarterly basis for a detailed breakdown of close to 60 industries.
Both developments will support a new quarterly growth accounting release from April 2018, which we hope will eventually provide MFP information at a lag of just 90 days after the period to which it relates for around 50 industries. This will move ONS from the very back of the timeliness pack to the very front and shift our level of industrial granularity from the very bottom, to around the middle of the pack.
While these improvements are considerable and will deliver a significant improvement, we recognise that there remains more to be done to transform our suite of productivity data to be comparable with among the best in the world. To this end, we plan to study the contents of this report in detail and to use its results alongside feedback from users to publish a Productivity Development Plan in April 2018. This will reflect on the advances made since the last development plan in July 2016.