简体中文
If the pandemic has been good for one thing, it's time in bed.
"People are taking a chance to get some extra sleep," says Daniel McNamara, a data scientist at energy supplier Bulb. "It's cool."
The insight comes from an unexpected source: smart meters.
These devices measure electricity usage and send the information straight to energy suppliers.
At Bulb, which has about 1.5 million customers, the change in morning behaviour (shown in the chart below) was very clear.
"Using the kettle, the microwave, turning the lights on, having a shower: that's the kind of thing you'd normally be doing at 7am to 7.30am if you were going into work," says Dr McNamara.
"Now that's been delayed until later, so we're seeing people cutting it a lot closer to the start of work. Around 9am is when we see usage picking up."
An extra hour or two in bed is not the only thing smart meters can spot.
They give us a glimpse into habits of households around the country - including whether or not they are occupied.
Here are the full results of the data shown to Sky News from its tens of thousands of smart meters, which broadly align to assessments of national representation breakdown published by energy regulator Ofgem.
At a moment when stopping transmission of new variants is crucial, they suggest the strength of lockdown - which is already weaker than the one in March - may be waning.
People are working at home less as the lockdown goes on
Electricity data can be used to track the effect of the pandemic on travel to work.
The technique is based on a simple observation: households who go to work use less electricity at lunchtime.
When we're at home, we boil the kettle or turn on the oven. When we're at work, none of this electricity usage takes place.
Using this method, Bulb divided households into two groups based on their weekday electricity consumption.
• Commuters, whose morning electricity peak exceeds their lunchtime peak
• Home workers, whose lunchtime electricity peak exceeds their morning peak
The chart below shows the percentage change in the number of commuters, compared with pre-lockdown levels.
Look at the end. Since lockdown started, the number of "commuters" identified by Bulb has drifted up - not definitively, but in a way that has the hallmarks of a trend.
This lockdown isn't as strict as the one in March
As you'd expect, the number of "commuters" fell sharply in late March and early April, before slowly recovering over the summer as Boris Johnson urged the country to "go back to work if you can".
During the second lockdown, when the rules on movement were looser, it was lower than normal, but barely dipped at all from the levels of the summer. This lockdown had little effect on workers.
The third lockdown is stricter than the second, but not as strict as the first. But why?
One may be that the definition of "key worker" has unofficially expanded: reports have suggested it is being used more often than in the past, perhaps because companies believe they have made their workplaces safe, but perhaps also because people are less willing or able to stay away from work.
Professor Paul Hunter of the University of East Anglia suggests another reason.
"The thing that really stands out is that the November lockdown, when schools were open, did very little to reduce working away from home," he says, especially in comparison with January.
"Perhaps the big impact of closing schools is not around preventing transmission within schools but rather around forcing more parents to work from home in order to look after their homebound children."
Londoners go to work less than the rest of the UK
Back in September, as cases increased in many parts of the UK, epidemiologists were puzzled by a question. What was happening in London?
Despite being home to some of the country's poorest and most densely packed communities, cases in the capital weren't rising.
Bulb's data suggests one possible answer: in London, more households are able to work from home. The chart below compares commuters in London with commuters in the rest of the UK.
As you can see, there are fewer in the capital - perhaps because it has more people working in jobs that can be done from home.
Of course, London has been hit very hard by both the first and second wave of the virus. Despite this, experts suggest it may have had some protection which other areas lacked.
"Results such as these add to evidence that London displays a greater level of adaption than, in this case, the UK as a whole, and that will have aided in the efforts to stop the spread of
COVID-19
in the capital," says Richard Harris, professor of quantitative social geography at Bristol University.
"It hasn't stopped it, of course (as people are still working and interacting). But, having fewer do so, helps."
There was a spike in "commuting" just before the second lockdown.
The day before the second lockdown came into force in England, movement jumped around the country. Numerous measures of movement data, like Google's measure of mobility, rose sharply.
We can now see that energy usage during the day dipped as well, suggesting that more people were out of the house.
After Sky News revealed this spike, Professor Hunter and colleagues linked it to a surge in new infections.
In terms of lives lost, the leak of lockdown was one of the most costly in UK history.
The picture is very fuzzy
These findings about the increasing numbers of people going into work pose an obvious question: if more people are spending time together, why are cases falling?
It is hard to say for sure, because it's possible that cases are not falling - or, at least, not falling fast.
The ONS survey suggests that they aren't, even though hospital admissions clearly are.
Government scientists have suggested that the difference may be the result of younger people, who would be less likely to end up in hospital, not coming forward for tests. In theory, this data could support that thesis. Are people avoiding tests in order to go to work?
We shouldn't rush to judgement, because the smart meter data misses a lot.
It would be easy to conclude that someone who ate a salad for lunch every day and didn't use the kettle was out at work.
Conversely, because it's at a household level, it might give the impression that someone whose children are now at home for lunch was in the house when actually they were out.
Right now, however, the overall picture is that, when it comes to leaving lockdown, workers are voting with their feet.
