Coronavirus, Scottish Economy, UK Economy

Weekly update – The GDP tales

When new GDP data come along, it can feel like be hard to know what they mean. But more than most, recent GDP releases have hit headlines as revisions have changed accepted narratives regarding how the UK has fared since the shock of the Covid-19 pandemic.

Why are there so many revisions?

Revisions to (almost all*) national statistics are a fact of life. This is because any attempt at economic measurement requires a trade-off between completeness and timeliness. It is important that we have a good idea about what is going on in terms of economic activity, and so we want to get timely readings. But some information takes time to become available, and so it’s natural that readings change when it’s more complete.

The ONS publishes output-based measures of GDP the month after the period in question, but some of the information such as business costs is not yet available. They then publish quarterly reading which has yet more information, but some administrative data such as self-assessed income tax and corporation tax returns isn’t available until well after the financial year has ended. At that point, the ONS reconciles other measures with the administrative information, which can cause significant revisions.

We could wait until all information is available. But doing so would be very costly. Businesses need information about economic conditions to plan their production schedules and investment decisions. Households need it to set their expectations of what might be coming down the line. The Bank of England uses economic data to understand what interest rates they could set. Governments need to know if the economy is performing well to know whether more or less fiscal action is necessary to achieve their objectives.

And so while immediate readings aren’t perfect, they are useful. But it would not be reasonable to ignore new information when it becomes available. And so that’s where revisions come in.

You may (if you’re really interested in the minutiae) have read that supply and use tables have been the source of the latest large revision. These are very detailed accounting tables of inputs and outputs by industry and product, and which allow the ONS to track how things are produced. Because of their complexity, they are only updated once a year. Normally this wouldn’t matter as much, but given the large movements in output during 2020 and 2021, changes have a larger effect than in normal times.

* Inflation measures are almost never revised due to ONS policy. For example, CPI has never been revised and would only be so in extremely exceptional circumstances.

Did UK GDP really bounce back faster than France and Germany’s?

One of the main talking points has been how the UK no longer has the slowest recovery from the pandemic of all G7 economies, and how it grew faster than France and Germany. But the reality is more nuanced.

Yes, if you take the latest available UK GDP data and plot it against the latest available French and German GDP, you would infer that the UK economy is now 1.8% larger than it was just before the pandemic, compared with 1.7% for France and just 0.2% for Germany; that UK GDP recovered its pre-pandemic level in quarter 4 of 2021, the same time as France and a quarter earlier than Germany; and that total growth since the trough in quarter 2 of 2020 has been 31.4% in the UK, compared with 23.5% in France and 12.4% in Germany.

But there are two important health warnings when doing these comparisons. One is that GDP measurement issues make it tricky to compare across geographies, and this is particularly true during the pandemic, when output in areas such as health and education was extremely hard to measure. For example, how do you capture the fact that during lockdown routine health interventions and education were so disrupted, given that there are no market prices for these in the public sector?

The approach chosen by the ONS meant that given that salaries paid to providers were unchanged, but much fewer health and education interventions took place, their implicit price increased substantially. This is best practice, and makes comparisons within country over time more meaningful – but most other jurisdictions produce measures based on inputs such as employment. Normally this doesn’t matter much – but the size of the shock during the pandemic brings this into sharp focus, although there is debate as to how much that has contributed to outcomes.

But the other health warning is one about timing. The ONS has been one of the first national statistics institutes to publish GDP estimates using revised supply and use estimates. But the Eurostat is due to publish their own revised estimates this Autumn, and there’s every likelihood that they will show a change in a similar direction.

The main takeaway? Too much focus on international comparisons, especially using what are fundamentally different vintages of data, is not particularly helpful and we can get too hung up on minor differences that can be revised away. What we know is that UK GDP today has recovered since the pandemic, but is barely higher than it was in 2019. And that is a challenging enough reality regardless of how we specifically compare with one country or another.

Scottish GDP news and the importance of the public sector to headline growth

A couple of weeks ago, the ONS released their first estimate of UK GDP for July, which showed a contraction of 0.5% relative to June. Health and education were two of the main contributors to the fall, and this was largely due to strikes in England. At the time, we highlighted that because of their absence in Scotland, we might see a very different story when the figures were released.

This has now been reflected in the statistics published by the Scottish Government, which showed GDP growing by 0.1% relative to June. Health and education both fell by 0.1% – but UK-wide the fall was 2.1% and 1.1%, respectively. Given the size of the public sector, especially in these two areas, their output can often be key to the direction of travel of the economy as a whole.

The wholesale and retail sectors fell more strongly in Scotland than for the UK as a whole, perhaps reflecting even worse weather in Scotland (who knew?!). But the real source of positive news was the manufacturing sector, which grew by 2.4% month-on-month, compared with a 0.8% fall for the UK as a whole – although output from manufacturing in Scotland is still slightly below pre-pandemic levels.


João is Deputy Director and Senior Knowledge Exchange Fellow at the Fraser of Allander Institute. Previously, he was a Senior Fiscal Analyst at the Office for Budget Responsibility, where he led on analysis of long-term sustainability of the UK's public finances and on the effect of economic developments and fiscal policy on the UK's medium-term outlook.