Anatomy of the euro area slowdown: lessons from the second half of 2018

What lies behind the slowdown?

The second half of 2018 was a cold shower for the euro area economy. After growing at a reasonable rate of 0.4% quarter-on-quarter in the first half of the year, growth fell to around 0.1%-0.2% in the second half. In this article, we analyse the cocktail of ingredients that led to this slowdown. This exercise takes on a particular relevance following the higher-than-expected growth in Q1 (0.4% quarter-on-quarter, 2 decimal points higher than expected): was this positive surprise temporary, or was what we witnessed in the second half of 2018 a mere bump in the road after which the economy will regain momentum in a sustained manner? To address this question, we take a detailed look at the role of three factors that have been behind the slowdown: the decline in car sales (in a context of difficulties in the automotive sector and poor performance in Europe’s manufacturing sector), the slowdown in international trade and the spike in political uncertainty.

In particular, we analyse the behaviour we expected to see from these three factors in the second half of 2018 under normal conditions and compare this with the actual data. Then, it is relatively easy to estimate the impact on GDP of the «deviations» that occurred for each of these factors.

Let us start with car sales in the euro area, which were affected by the regulatory uncertainty caused by the new regulations for vehicle emissions that came into force on 1 September 2018 throughout the EU. We estimate that the growth in new passenger car registrations in the euro area was 4.3 pps lower than expected, which subtracted 0.1 pp1 from quarter-on-quarter GDP growth.

With regard to the slowdown in international trade in a context characterised by protectionist tensions, the growth of exports was 0.4 pps lower than expected. This reduced GDP growth in the second half of the year by 0.09 pps (this estimate takes into account the import content of exports).2 Of course, part of the slowdown in exports can be explained by the lower exports of vehicles (we estimate that this accounts for slightly less than half of the slowdown), held back by the cutbacks in production to adapt to the new regulations and by lower global demand.

Finally, political uncertainty also played an important role, affected at the international level by the geopolitical and trade tensions and at the domestic level by the tensions between the new Italian cabinet and Brussels and by Brexit. In particular, we compare the trend observed in the index of political uncertainty in Europe produced by the US academics Baker, Bloom and Davis with the trend that was expected based on the index’s historical series. The observed trend in the uncertainty index proved to be 52 points higher than the expected trend (see second chart), a significant difference if we consider that 80% of the central observations lie within a range of154 points. We estimate that this factor eroded growth by 0.05 pps.3

If we bring all the results together (see third chart), we see that the three factors together reduced the average quarter-on-quarter growth of the euro area by 0.24 pps in the second half of 2018, a figure that represents practically the entire slowdown observed. In particular, 40% of the slowdown can be attributed to the drop in vehicle sales, slightly more than a third to the slowdown in trade and one fifth to political uncertainty.

Besides this breakdown, another way to characterise the slowdown in the euro area is to ask ourselves whether it is due to supply factors (i.e. factors that restrict the economy’s productive capacity or potential, such as the appearance of bottlenecks) or demand factors (such as households’ willingness to consume). This distinction can also provide clues as to how long the moderation in growth could last. On this note, a detailed quantitative study by the ECB4 that focused on manufacturing activity estimates that the slowdown has mainly been caused by weak demand, although the constraints on the supply side have played a more minor role (see fourth chart).5

Lessons for the present and the future

This «anatomical» exercise offers us some clues about the economic outlook for the euro area in 2019. With regard to the adjustment observed in the automotive sector, as vehicle manufacturers fully adapt to the 2018 European regulations and consumers know where they stand, vehicle sales can be expected to return to normal. In fact, this is already reflected in the data for the first quarter: the number of vehicles registered has increased by 7.0% quarter-on-quarter, after falling by 11.0% in Q4 2018. However, all things considered, regulatory uncertainty is likely to continue to affect the sector: the EU is preparing new and stricter emission standards that will need to be met by 2021 and more European cities are expected to introduce restrictions on the circulation of vehicles.

As for the slowdown in trade, the balance is mixed: on the one hand, an appeasement in the conflict between China and the US could help to boost trade flows, although there is a risk of the US administration imposing tariffs on European cars. On the other hand, the slowdown in the European economies’ main trading partners, in combination with a more uncertain global environment for global trade (less confidence, higher non-tariff barriers and greater pressures on global supply chains), suggests that the foreign sector will continue to make a modest contribution to euro area growth.

Finally, political uncertainty is likely to remain high and further spikes cannot be ruled out if there are surprises in the European elections, with Brexit or if other sources of tension, such Italy’s fiscal policy, are revived. That said, a source of hope is the fact that in Q1 2019 – a difficult period dominated by the threat of a chaotic Brexit – the political uncertainty index in Europe decreased by 24 points compared to Q4 2018, mostly thanks to a softening of tensions in France and Italy.

Javier Garcia-Arenas

1. We compared the observed growth with that predicted by an AR(1) model based on historical evidence. We then multiply the deviation obtained by the relative weight of the consumption of cars in the total consumption of European households (4%). Finally, to estimate the impact on growth, we take into account the fact that private consumption accounts for slightly more than half of the euro area’s GDP.

2. A year ago, at CaixaBank Research we predicted that the quarter-on-quarter growth of exports in Q3 and Q4 2018 would be 1.2% and 1.1%, respectively, but the observed growth was 0.2% and 1.2%. We multiply the deviation in export growth by its weight in the euro area GDP (28%), subtracting the import content of exports (25.7% according to the OECD).

3. Specifically, we compare the observed values of the index with those predicted using an AR(1) model. Then, we plot a regression of GDP growth against the index and we make a prediction for GDP growth in the second half of the year using the observed values and those obtained with the AR(1) model.

4. See M. Dossche and J. Martínez-Martin (2018). «Understanding the slowdown in growth in 2018». ECB Economic Bulletin, Issue 8.

5. Other studies also come to the same conclusion. See, for example, A. Benito and P. Vernet (2019). «European Daily: Anatomy of a slowdown». Goldman Sachs Economic Research.