COVID-19 and social distancing: does it show up in the demand for electricity?

By Steven Percy, VEPC Senior Research Fellow, Kelly Burns, VEPC Senior Research Fellow, and Bruce Mountain, VEPC Director

Empty motorways and airports make it obvious that movement restrictions and stay-at-home regulations have had a big impact on the demand for oil and petroleum products. But what impact have these restrictions had on electrical demand? We examine electrical demand data measured at high voltage transmission grids to assess the impact of Covid-19 social distancing restrictions in Australia, the United States, New Zealand and Great Britain. Interestingly, New Zealand and Australia have had amongst the lowest per capita infection and case fatality rates, while the United Kingdom and United States have had amongst the highest.

We also review changes in mobility as measured in Google’s Covid-19 Community Mobility Reports(1). We find a strong correlation between mobility trends and aggregate electrical demand. While apparently similar social distancing restrictions in all four countries might have been expected to show up in similar electrical demand and mobility reductions, in fact the picture is very different: electrical demand (and community mobility) declined sharply in New Zealand and the UK. In Australia and much of the US, electrical demand has hardly changed. Though mobility reduced in the US and Australia, the reduction in both countries has been much smaller than in New Zealand and the UK.

(1): Community Mobility Reports aim to provide insights into what has changed in response to policies aimed at combating COVID-19. The reports chart movement trends over time by geography, across different categories of places such as retail and recreation, groceries and pharmacies, parks, transit stations, workplaces, and residential. The reports can be found at: https://www.google.com/covid19/mobility

Background

The UK, New Zealand and Australia introduced social-distancing regulations on 23 March, 23 March and 31 March respectively. In the US these regulations were established by state governments in California, New York, Florida and Texas on 19 March, 22 March, 30 March and 30 March respectively. Some of the other US states introduced similar restrictions around these dates. Contemporaneously, border controls and quarantine for international arrivals were imposed in New Zealand, Australia and the US, but not in the UK.

Change in electrical demand

Figure 1 shows the country-level changes in aggregate electrical demand compared to a historical baseline(2). Electrical demand in both the UK and New Zealand declined significantly after social distancing regulations were imposed while demand has been largely unchanged in Australia and has a decline a little in the US, relative to the baseline.

(2): These were calculated by extracting the trend component of the 30-minute aggregate energy demand from 2017 to 2020 using seasonal and trend decomposition (STL) applying locally estimated scatterplot smoothing (Cleveland et al., 1990). The 2020 energy demand trend is compared to the average of the demand trends from 2017 to 2019. No temperature adjustment is applied.

Structural differences cannot account for the differences we observe in country-level changes in aggregate electrical demand since COVID19 as depicted in Figure 1. According to the International Energy Agency (World Energy Balances, 2019), the amount of electricity consumed by the commercial and public service sectors relative to the residential sector is largest in Australia (followed by US and UK) and lowest in NZ. If COVID19 restrictions had resulted in the same proportional decline in electricity consumed by the commercial and public sectors across countries, aggregate electrical demand would fallen more in Australia (followed by US and UK) and least in NZ. But we see the opposite thus we conclude that structural differences in the productive sector of the economy cannot account for the country-level difference in changes in aggregate electrical demand compared to the baseline since COVID19.

Figure 1: Trend change in aggregate country demand relative to baseline

Source: USA: EIA (US Energy Demand by Region); NZ: Electricity Authority (Latest wholesale trends); UK: National Grid (Data Explorer) and ELEXON (BM Reports), AU: AEMO (NEMWeb). VEPC analysis.

Figure 2 shows the weekly average change in electrical demand compared to a historical baseline for New York, Florida, Texas and California, and Figure 3 shows the same measure in the Australian states of Victoria, New South Wales, South Australia, Tasmania and Queensland.

Figure 2: Percentage change in average weekly demand in selected U.S. States, relative to baseline

Source: EIA (US Energy Demand by Region). VEPC analysis.

Figure 3 shows the largest demand reductions in New South Wales, almost no change in Queensland and Victoria and slight increases in Tasmania and South Australia. It is also obvious in these charts that, relative to the baseline, the variation in demand before the social-distancing regulations took effect was at least as large as it has been since.

Figure 3: Percentage change in average weekly demand in selected Australian states, relative to baseline

Source: AEMO, NEMWeb. VEPC Analysis Figure 2 shows the largest demand reductions occurred in California and New York and that demand increased in Texas and Florida.

Change in mobility

The direction of the country and regional demand changes since the introduction of social distancing regulations is consistent with the Google mobility data shown in Table 1. The table shows the average percentage reduction in mobility since the imposition of social distancing regulations.

Table 1: Average percentage change in mobility since social-distancing regulations

Source: Google Covid-19 Community Mobility Report, VEPC analysis

Table 1 shows that, consistent with the reductions in electrical demand, the biggest reductions in mobility occurred in New Zealand, the UK and in California and New York. The smallest reductions in mobility occurred in Australia (approximately the same reduction in all states) and in Texas and Florida.

This analysis suggests that changes in mobility as measured at retail & recreation venues and workplaces, and changes in aggregate electrical demand are strongly associated and that average reductions in mobility above 45% are associated with large (10% plus) declines in electrical demand, but reductions in mobility below 45% are associated with little change in electrical demand.

Of course the specific circumstances (particularly climate), the choice of baseline and the economic structure of economies affect demand. A rigorous economic analysis will unearth additional features and can isolate more precisely the impact of social isolation policies. However this indicative analysis suggests that in developed economies, reductions in social mobility and electrical demand are strongly associated but that reductions in mobility of less than 45% in workplaces and recreational spaces have had little impact on aggregate electrical demand relative to the baseline. Reductions above this level (in New Zealand, the UK, New York and California are associated with reasonably large (10% plus) reductions in electrical demand relative to baselines.