Revisions to estimates published in the annual Blue Book 2020 were relatively modest in comparison to previous years, and revisions up to three years after the first published estimate are not statistically significant.
Revisions are typically larger around turning points, but this might be more pronounced because of the challenges in compiling their National Accounts over the course of the pandemic.
There is a trade-off between the timeliness and accuracy of estimates of gross domestic product (GDP). As additional information becomes available, we have a more complete picture of economic activity in that period. This production cycle can take up to three years and leads to revisions. The annual Blue Book process is also when major methodological improvements are introduced.
To assess revision performance, we can estimate:
the mean revision (MR), which shows whether there is a systematic tendency for initial estimates to be revised upwards or downwards
the mean absolute revision (MAR), which measures the absolute size of revisions so that upward revisions are not offset by downward revisions of the same magnitude
the mean square revision (MSR), which incorporates the degree of bias and the variance of the revision, as large revisions are treated more seriously than small revisions
This article analyses the revisions in Blue Book 2020. It then looks at initial revisions to UK GDP compared with those in other countries in 2020.Back to table of contents
Figure 1 shows the revisions to volume estimates of gross domestic product (GDP) in recent Blue Books. Revisions to Blue Book 2020 were relatively modest with a mean absolute revision (MAR) of 0.08 percentage points compared with an average of 0.11 percentage points over this entire period.
Figure 2 shows there is a zero mean revision (MR) at T+3 months, implying that there is no tendency for the first estimate to be revised up or down. The MR is a little higher when comparing the first published estimate with later ones. The initial revisions (T+3 months) and those that tend to reflect the incorporation of annual benchmarks and balancing in a supply and use framework (T+24 months and T+36 months) are not statistically significant (Table 1). This is not the case for those revisions in the more mature estimates. Since these methodological changes cannot be anticipated at the time of the first estimate, these revisions are a less accurate reflection of the quality of early estimates.
Figure 2 shows that it is more likely that there will be a higher frequency of offsetting revisions in the first two years, relative to the subsequent revisions. This highlights why a low MR can be a misleading signal of quality. While the MR is largely similar at T+3 months and T+24 months, there is a more marked increase in the MAR.
|T + 3||T + 24||T + 36|
|Timespan||Mean Revision (pp)||Absolute Average Revision (pp)||T Score²||Statistically Significant?||Mean Revision (pp)||Absolute Average Revision (pp)||T Score²||Statistically Significant?||Mean Revision (pp)||Absolute Average Revision (pp)||T Score²||Statistically Significant?|
|1961 Q2 to 2017 Q4 1||0.01||0.06||0.9361||No||0.04||0.50||0.8706||No||0.10||0.53||1.9408||No|
|1961 Q2 to 1969 Q4 1||n/a||n/a||n/a||n/a||-0.01||0.69||-0.0606||No||0.06||0.79||0.3830||No|
|1970 Q1 to 1979 Q4 1||n/a||n/a||n/a||n/a||0.08||0.91||0.4800||No||0.20||0.98||0.9649||No|
|1980 Q1 to 1989 Q4||0.03||0.06||1.3791||No||0.09||0.80||0.5539||No||0.18||0.71||1.2704||No|
|1990 Q1 to 1999 Q4||0.01||0.07||0.4767||No||0.06||0.18||1.7550||No||0.10||0.22||2.5717||Yes|
|2000 Q1 to 2009 Q4||0.00||0.11||0.0659||No||0.01||0.23||0.1981||No||0.00||0.27||-0.0817||No|
|2010 Q1 to 2017 Q4||0.01||0.07||0.7896||No||0.00||0.15||0.1405||No||0.05||0.15||1.6639||No|
Download this table Table 1: Revision information and T-Test for statistical significance for quarterly GDP growth.xls .csv
Figure 3 shows that the quality of early estimates has improved over time, which likely reflects cyclical and structural factors. There have been improvements to the measurement of GDP as well as a lower degree of volatility in the UK economy. Figure 4 shows that the earlier periods were inherently more volatile for the UK economy relative to the most recent periods, which has also coincided with there being lower revisions of late - although this excludes the coronavirus (COVID-19) experience for 2020 onwards. There have also been improvements in how we record GDP estimates, which is likely to be another explanation for the lower mean square revision (MSR) in these years.
Back to table of contents
Gross domestic product (GDP)
A measure of the economic activity produced by a country or region. There are three approaches used to measure GDP:
- the output approach
- the expenditure approach
- the income approach
A more detailed glossary is available.Back to table of contents
Revisions are typically larger around turning points. It has been particularly challenging for National Statistical Institutes to compile gross domestic product (GDP) estimates over the course of the pandemic, given the heightened levels of statistical uncertainty.
As more information becomes available over time, this might lead to higher levels of revisions in more mature estimates, which might also impact upon international comparisons. We will communicate such revisions as part of our future engagement to improve wider understanding of the effects of the coronavirus (COVID-19) pandemic.
Revisions in this article relate to data published in the UK National Accounts, The Blue Book: 2020.
More detailed information on the compilation of the latest national accounts, including quality and reliability issues, is available in the background notes.Back to table of contents
Contact details for this Article
Telephone: +44 (0)1633 455284