Macro insights

Assessing the service sector recovery

Mobility is starting to recover as governments ease lockdown restrictions. How might this impact economic activity? Based on an analysis of developed market countries and our earlier GDP shock-calibrations, we arrive at three main findings.

Sources: BlackRock Investment Institute, Oxford University and Google with data from Haver Analytics, June 2020. Note: This chart shows the impact that a 10 point move up or down in consumer mobility data and stringency measures have on the service sectors of nine developed market economies. We use google data on retail and recreation mobility for the mobility score and Oxford data on lockdown stringency. We use a panel regression technique that combines country and time dimensions to measure the marginal effect of mobility on service sector activity, and how this differs between tightening and easing phases.

First, service-sector activity is impacted more by measures of mobility than the severity of lockdown measures, the service sector may recover more slowly than it contracted, and services have been hit harder than manufacturing. Second, a gradual recovery in mobility in early June may not provide a further significant lift to June purchasing managers’ indices (PMIs). Third, GDP will likely bottom out in the U.S. and euro area in the second quarter of 2020. Social distancing measures may be relaxed at a slightly faster pace in the euro area and therefore weigh less on GDP than in the U.S. That said, this difference will likely be outweighed by stronger US policy support.

Our results on the relationship between the Purchasing Managers’ Indices (PMIs), lockdown stringency and mobility patterns suggest that mobility data have a stronger impact on business sentiment than the lockdown measures. For example, a 10-point increase in stringency is associated with a 4 point decline in the services PMI. But a 10-point drop in retail and recreation mobility is associated with a 5.2 point decline when measured from the same starting point. See the chart above.

The speed of service sector recovery as lockdowns are eased seems to be slower than the speed of contraction when lockdown measures were imposed across DM countries. In other words, the relationship is asymmetric between tightening and easing. A 10-point tightening in mobility led to a 5.2 point decline in the services PMI, yet a 10-point easing only recovered 3.7 points points (see the difference between the red and yellow bars in the chart above).

The service sector was hit harder than the manufacturing sector by the sudden stop in activity. An equivalent drop in mobility or stringency leads to a much sharper decline in the services PMI than for manufacturing. This is intuitive, given that services tend to be more contact-intensive and that in many countries the manufacturing sector was able to operate even at the height of lockdowns.

Sources: BlackRock Investment Institute and Google, with data from Haver Analytics. Notes: The chart shows the services purchasing manager’s index (PMI) for the U.S., Germany, France, Italy and the UK and the Google consumer mobility scores (data for retail and recreational activity) for those countries. June estimates are based solely on mobility trends.

We can use these relationships to arrive at preliminary estimates for the June PMIs assuming that mobility continues to normalize gradually. The result of this exercise suggests only a slight pick-up in June manufacturing and services PMIs relative to May. See the chart above. It is important to note that these estimates are based solely on mobility trends. Yet they do illustrate that – in the absence of a policy response and changes in behavior – there is a high bar for activity to quickly recover to pre-coronavirus levels. 

Finally, we can use the severity of social distancing measures and a simulated path for their easing to map out a longer-term path of economic activity. This part of the analysis is based on the shock calibrations we highlighted in May. Assuming a gradual easing in social distancing measures, GDP growth would rebound after a steep contraction in Q2 2020. Given the slightly earlier and more rapid normalization in social distancing and the more favorable virus dynamics in the euro area, the pickup in GDP implied by the change in social distancing is a touch quicker than in the U.S. But once we factor in the fiscal and monetary policy stimulus that has been put in place over the last few weeks this differential is likely to be reversed, given the stronger stimulus in the U.S.