Tuesday, 3 July 2012

How we use electricity at home - a survey

DeFRA have recently published a survey of electricity usage [1] in a sample of 251 households of various kinds. Intertek, the firm that conducted the study, metered every fixed appliance: lights, TVs, washing machines, hair dryers, smoke detectors, electric blankets – if it was possible to meter it they did. Most of the households took part for one month but 26 households were recruited for a full year which showed seasonal variation in heating, lighting and so on. The report includes a huge amount of data, some surprising and some less so. Here is just a selection of what I found most interesting.You can download the full report from here.

Total electricity use

Ignoring households which used electricity for heating, the most profligate were multi-person households without children whereas families used 11% less than average. Pensioners living alone were most frugal but pensioners living together used considerably more. This chart shows household types by increasing electricity use.

Average annual electricity consumption by household type (explained in Notes below)

NB Over all households, the variation was much greater:  at least 10%  used more than 6000 kWh/year and another 10% used 1500 kWh/year or less.

Through the day, the minimum load was on average about 170 W at 4am. The peak was 600 W between 5pm and 8pm. Less than half of the minimum load was from fridges and freezers. It is difficult to say how much of the rest was unnecessary standby current.

Winter peak load

Reducing peak load is important for reducing the cost of backup for intermittent renewable energy (see Smart meters can save us money and How to balance electricity supply and demand). The biggest peak for the UK is early evening in winter and this survey reveals how heating, cooking and lighting contribute to that peak. Households using electricity for their main heating use very little at peak time but households using electricity as additional heating use more in the evening that at other times -  though it still isn’t more than about a fifth of overall load. For households without electric heating the peak load is 30% cooking, 25% lighting, and another 45% coming from a range of appliances. However, the increase at peak time is mainly due to cooking and lighting. The lighting component could certainly be reduced if we used more low energy bulbs which brings me onto that subject.

Winter electricity load for households without electric heating: from [1]


The average household has 34 lamps of which 13 are incandescents and only 11 are low-energy bulbs (CFL or ordinary fluorescents). We are most likely to use CFLs in office and hallway but even then only about a third. In the lounge, dining room and bedrooms about half the bulbs are still incandescent.

In terms of overall lighting energy, the average household used 537 kWh/year on lighting and the most frugal type were pensioners living together. Lone pensioners used more than the average – my guess is that they are leaving lights on to deter burglers.  Families also used less than average.

Average annual electricity consumption for lighting

Across all households, many used less than 100 kWh/year but more than 1000 kWh/year was not unusual.

TVs and audio visual

The 251 households owned 417 TVs so 1.7 per household on average. The audio visual 'site' comprised the TV and also set top box, DVD players, games consoles etc.

Average annual electricity consumption for TV and accessories.

Across all households the range was staggering, with 10% using less than 200 kWh/year and a few using more than 3500 kWh/year. The highest energy TVs were plasma: 246 W on average, compared to 97 W for an LCD TV.

Pensioner + households watched the least and other adults living together watched the most, more even than families, who you might expect to be the biggest addicts.

Households with plasma or LCD TVs had them on for an average of 5.5 hours/day. Old-style CRT TVs were used a little less: only 4.2 hours/day.

The average standby power usage for an LCD TV was 1.7W and Plasmas only 0.5W. However the average standby power for the whole audio-visual system was 17.8 W.

This report did not find any difference in TV watching through the year. This is probably because the sample was too small to be statistically significant. BARB [2] who survey viewing figures for all the TV channels find we watch about 6% more than average in Jan/Feb and 6% less in May. Which brings me to next to seasonality.


Lighting varies through the year: we use 3 times as much energy for lighting in mid winter as in the summer..

Cooking varies too albeit much less: 40% more in winter than summer. (Barbecues were  not in the survey - see How green is my charcoal barbecue)

Fridges and freezers are the opposite, using 50% more in the summer than in the winters.

Laundry varies as well: 65% more energy used in winter than in summer, presumably because we are using the free sunlight to dry clothes in the summer, or maybe because summer clothes weigh less and take less energy to wash?



The report measured the power consumption of the computer with all its peripherals together so printer, modem etc. are all in the same bucket. This makes the on/standby/off data confusing – for example when my site is idle the monitor goes into standby but the computer does not. From the average daily load curve, a lot of households leave computer equipment ‘on’ all night, especially multi-person households without kids - for them the the minimum consumption was around a half of the peak consumption. Families and lone pensioners were best at turning off but even for them the minimum consumption was about a third the maximum.

Even lone pensioners used an average of 137 kWh/year at the IT site, compared to a household average of 240 kWh/year. Many lone pensioners switch off for a while around 3pm especially on holidays. There is nothing corresponding to this in the TV graphs. Is there something on the radio? If the 3pm value is actually 2-3pm maybe it is the Archers.
Average annual electricity consumption for computers and peripherals
Load curve for computer and peripherals, lone pensioners on holidays [1]

Fish and reptiles

The equipment for looking after fish needs a lot of energy, reptiles rather less. On average:
kWh/yearFor comparison:kWh/year
aquarium278 refrigerator16
pond pump218 dishwasher294
vivarium 57 upright freezer372

Other appliances

8 of the 251 households had no kettle – presumably they used the hob.
5 households had electric doorbells. These took a whopping 52 kWh/year.
One household had an electric chair. This was a recliner, not a device for executing relatives. It used 13 kWh/year.


Intertek used Ipsos MORI26 to select households to be as representative of the general population as possible. However, they had to exclude tenants due to the complexity in getting permission which explains why there were fewer students than the national average. Also there were more retired people and less full time workers. Finally the households which completed the study, although they fit the national profiles for age and household type, were more ‘green’ in attitude than average: they were more likely to be ‘positive green’ or a ‘waste watcher’ and less likely to be  ‘cautious participant’, ‘sideline supporter’, ‘stalled starter’ or ‘honestly disengaged’. Presumably this will have some impact on energy use behaviour – though it is difficult to say how much.

The household types are:
  • Lone pensioner: one person aged 65 or more
  • Lone non-pens.: one person under 65
  • Pensioner +: more than one person, at least one aged 65 or more
  • Multi with-kids: more than one person, at least one under 18
  • Multi without-kids: more than person, none under 18

[1] Household Electricity Survey: A study of domestic electrical product usage, May 2011 DeFRA

[2] Broadcasters Audience Research Board Monthly total viewing summary, May 2011-May-2012


  1. In our household, the main cause of the increase in our electric bill is the use of the television 24/7. That’s why I have formulated a schedule of when the TV is to be opened to save energy. Same goes with the video games and our desktop computer.

  2. Did you know that there are TVs now which automatically turn the screen off when no-one is watching? Our one does. I think it may be just Sony that does this. It uses an infra red presence detector. However, if there really is someone watching 24x7 this won't help very much.