22 Feb 2017

UK Q4 GDP (2nd estimate): Q4 growth stronger, but warnings for 2017?

Philip Shaw

Chief Economist

Quarterly GDP growth in the last three months of last year has been revised up to +0.7% q/q, from the initial estimate of +0.6%. The median consensus view was for no revision, but several economists (including ourselves) had expected the uplift. However, downward revisions to the back data have taken the shine off the Q4 upgrade – calendar year growth has been revised down from 2.0% to 1.8%.

Drilling into the detail of the revisions, the uplift to the Q4 growth estimate came from the production and construction sectors, as updated output estimates for these sectors were incorporated in the GDP numbers. Industrial production growth has been revised up from 0.0% q/q to 0.3%, while construction output growth was nudged up from +0.1% q/q to +0.2%. Growth in the all-important services sector was confirmed at a robust +0.8% q/q. Prior to Q4, the biggest downward revision was to growth in Q1 last year, which has been shaved down to +0.2% q/q from +0.3% estimated previously. All told, the revisions mean that the level of Q4 GDP is 0.14pp lower than previously estimated. But the picture of post-EU referendum momentum in the economy now looks a touch stronger than before – across the four quarters of the year, 2016 saw GDP growth of 0.2%, 0.6%, 0.6% and 0.7%.

Today’s release also provided the first steer on the expenditure breakdown of Q4 GDP growth. From these data, we think that, upward Q4 GDP growth revision notwithstanding, there are three (tentative) signs that a growth slowdown in prospect.

  • First, business investment declined by 1.0% q/q (delivering a fall of 0.9% y/y). While these data are volatile, that would be consistent with Brexit-related uncertainty weighing on business spending.

  • Second, while there was a 1.3pp boost to Q4 GDP growth from net trade (exports up 4.1% q/q, imports down 0.4% q/q), this outturn looks to have been significantly flattered by data quirks. Specifically, the ONS pointed out that there was a large boost to net trade from outflows of non-monetary gold holdings from the UK. That effect is GDP neutral as it was offset, in the ONS’s calculations, by a huge downswing in the ‘acquisitions less disposals of valuables’ category. This category, which usually only makes tiny contributions to GDP growth, is estimated to have detracted 1.4pp from the Q4 growth rate. Stripping out the gold effect, the ‘underlying’ net trade boost might have been minimal – that is disappointing, given the extent of the fall in the pound we saw last year.

  • Third, we note that the economy needed to rely heavily on household consumption in Q4 – this category of spending rose by 0.7% q/q. But we note that, with the rate of inflation rising (predominantly as a result of the weaker pound and higher import costs), indicators of household spending are beginning to turn southwards – our reaction to last week’s soft retail sales numbers provides more detail (‘Consumers start to feel the squeeze…’).

The confirmation of strong overall growth momentum towards the end of last year is, of course, encouraging. But we reckon that headwinds to business and, in particular, household spending will blow harder this year. And we question whether a sterling-related boost to net trade completely offset these headwinds. For that reason, we maintain our view that the economy will slow down, albeit modestly, over the next year or two.

The MPC shares this view. The committee also believes there is a degree of ‘slack’ still present in the economy (it now believed the ‘equilibrium’ rate of unemployment is 4.5% - lower than the current rate of 4.8%). Granted, the MPC, like us, reckons that CPI inflation will rise well above the 2% target over the next couple of years as sterling effects on import prices feed through to consumer prices. But to the extent that the increase proves only temporary, policymakers will continue to look through them. If the economy does slow down as we expect, we do not see a hike in Bank rate until at least the second half of 2019.