Our weekly index suggests that activity dropped from 100 in February to a low of 44.4 by end-April with the imposition of the lockdown.
By Sonal Varma & Aurodeep Nandi
India has finally embarked on Unlock 1.0, drawing to a close four phases of lockdown starting from March 24. There is likely to be a considerable gap between the theoretical opening up of the economy, and the actual resumption of economic activity.
Lockdown relaxations in India are being phased out in three stages. Unlock 2.0 in July will extend to educational institutions, while Unlock 3.0 (the timeline remains unspecified at this time) will see the resumption of international travel, metro trains, cinema halls, sporting facilities, assembly halls, and allow for larger congregations.
Assessing the pace of return to normalcy remains crucial in understanding the hit to growth and the expected pace of recovery. Ideally, the accurate way to assess this is to track the capacity utilisation. However, capacity utilisation surveys typically come with considerable lag and pertain to the manufacturing sector.
In the absence of updated capacity utilisation measures, we use a simplified version of the ‘Wharton Measure of Capacity Utilization’ to arrive at a “proxy” measure, ie, Normalisation Index (NI). Now, capacity utilisation rate is simply the ratio of actual output to total capacity. We use monthly data from 2000 onwards, up to May 2020 for our analysis. Then, we estimate the NI for each month as the ratio of the actual output (in seasonally adjusted level terms) in that month to the ‘peak output’. For our analysis, we consider a range of monthly high frequency indicators across the private demand side (consumption, investment and external sectors) and the private non-agriculture supply side (industry and services). We have applied the NI concept to a broader range of (non-industrial) indicators.
Note that there are some drawbacks to this methodology, the chief among them being that it assumes the historical peak necessarily represents full capacity. Our methodology of computing “proxy” full capacity should suffice as a rough indicator.
Consumption indicators, on average, saw a steep decline. Similar to consumption, both investment and external sector indicators witnessed a steep fall during March and April, although marginally lesser than the hit to consumption. Early data for May, such as railway freight traffic, has risen by 11.3pp from April to 69.4%. For the external sector, we see NI rise by ~34pp to 59.5%.
On the supply side, we estimate that the NI in the industrial sector dropped from 81.3% in March to 43.9% in April and is on track to rise to slightly over 70% in May. We estimate that services NI dropped from 81.4% in February to an average of 42.5% in March and a dismal 3.3% in April. Early data for May, such as commercial vehicle sales , suggest a rise to 6.7%, implying a much slower pace of activity normalisation in services sectors.
We estimate that activity normalization is occurring much slower in our measure of aggregate demand vs aggregate supply. Prima facie this is not surprising given that the larger hit from the lockdown typically happens to demand, while the supply side is constrained only to the extent mandated by the rules.
The above analysis is useful to analyse the extent to which activity remains below peak capacity levels, these ‘conventional’ indicators are only available at monthly frequency. To capture more recent trends, we capture weekly data across a host of ‘unorthodox’ indicators.
A useful tracker of the status of lockdown in a country are the indices compiled by the University of Oxford, through the Oxford COVID-19 Government Response Tracker (OxCGRT). We find that stringency was relaxed incrementally in end-April, with the major relaxation in the first week of May, when it dipped by 15%. Between May and June, the relaxation has been more benign, amounting to close to 8%.
Google’s Covid-19 Community Mobility Reports are a useful tracker of the variation of economic activity. We get a more up-to-date capture of trends using Apple Maps’ Mobility reports. Mobility still remains 60pp below pre-pandemic levels in January.
State-wise trends show that mobility has a broadly positive correlation with severity of the pandemic, i.e., states with higher cases of Covid-19 have experienced lower traffic levels.
Other indicators are power demand. As of the week ending June 14, power demand has somewhat recovered to levels seen in March. Labour market: According to CMIE, the unemployment rate had risen to 23.8% for the week ending March 29. Consumer sentiment: Consumer sentiment data suggests a much more muted recovery.
We have standardised each of these series by indexing their values for week ending 23 February to 100 (approximately a month before the lockdown was announced) and estimated a simple average. Our weekly index suggests that activity dropped from 100 in February to a low of 44.4 by end-April with the imposition of the lockdown. Since the economy started opening up through May, activity has recovered, with the provisional value for NIBRI at 68.5 as of 14 June, ie, still 31.5pp below the pre-lockdown levels of business activity.
This underlines our conviction that Q2 is likely to be a washout. We expect GDP growth to plummet from 3.1% y-o-y in Q1 to -15.2% in Q2. Beyond Q2, as lockdowns are relaxed, we expect activity to resume and the labour market to continue to mend. Evidence from South Korea and China suggests that the reopening phase releases some pent-up demand for durable goods (autos, furniture) and housing.
The reason this may be unsustainable beyond the rise in pent-up demand could be threefold. First, India has struggled to flatten the infection curve despite the lockdown, with the number of cases growing by nearly 4% day-on-day. As lockdown deepens, the curve is likely to become even more unwieldy, which could translate into lower business and consumer sentiment and activity. Second, the government’s fiscal support package has been relatively modest at ~0.8% of GDP, with much of it focused on financial and regulatory easing to alleviate immediate cash flow related problems. While this may limit the ‘survival stage’ shock to firms and households, this is unlikely to boost demand. Third, the Covid-19 shock may amplify the balance sheet weakness. Both erstwhile stressed, as well as healthy companies are currently struggling under a hit to profitability, which could lead to either a combination of cost savings and re-leveraging, or in more dire cases, insolvency. This could have a domino effect on jobs, which will either be lost, or will pay less, in turn impacting consumption demand. Stressed balance sheets of corporates and households will weigh on banks and shadow banks, which could tighten lending standards amid financial stability concerns.
Consequently, we expect GDP growth to remain in negative territory for the rest of 2020 and into Q1 2021 – projecting growth to recover from -15.2% y-o-y in Q2 to -5.6% in Q3, -2.8% in Q4, and -1.4% in Q1 2021. Overall, we expect growth to average -5.0% in 2020 and -6.1% in FY21.
Edited excerpts from Asia Insights, Nomura (June 15, 2020)
Authors are research analysts with Nomura Holdings. Views are personal