Family spending in the UK: financial year ending 2017

• Average weekly household spending rose to £554.20 in the financial year ending 2017; in real terms, this was a return to pre-economic downturn levels.

• Households spent an average of £79.70 a week on transport, an increase of £5.40 in real terms when compared with the previous year; this makes transport the top spending category.

• When comparing between English regions and UK constituent countries, average weekly household spending differed by more than £200.

• Households in the 65-to-74-year-old age group spent nearly a fifth of their total spending on recreation and culture.

• Households without children spent a higher proportion of their total spending on transport than households with children.

• Average weekly spending of a 15-year-old was more than three times that of a 7-year-old.

• We are celebrating 60 years of Family Spending – thank you to the 375,000 households that have taken part in the Living Costs and Food Survey and its predecessors since 1957.

Figures in this bulletin are averaged across all UK households, unless stated otherwise. All spending estimates are rounded to the nearest £0.10.

This bulletin uses the mean when referring to averages unless stated otherwise. Therefore, total average weekly household expenditure is equal to the total weekly expenditure of households divided by the number of households.

Results presented in this bulletin cover the financial year ending (FYE) 2017; that is, April 2016 to March 2017. It therefore contains data for time periods before and after the polling day for the EU referendum.

Spending is presented using classification of individual consumption by purpose (COICOP) categories, unless otherwise stated. COICOP is an internationally recognised classification system consistent with that used by UK National Accounts. It does not include all types of payments; for example, capital mortgage repayments are excluded. Sections 4 and 5 of this bulletin reference housing costs outside the COICOP classification system; this is to give a more complete view of this topic area. Estimates for the 60 years time series analysis do not use the COICOP classification system to compare over this longer time series. Further information on COICOP can be found on the United Nations Statistics Division website.

This bulletin compares spending over time in two different ways, where figures either are or are not adjusted for inflation. Where figures are not adjusted for inflation, differences in spending could be because of differences in price level between the years. When comparing spending that has been adjusted for inflation, differences in price between years are not the cause for differences in spending, as this has already been considered. Where possible, we adjust for inflation when comparing levels of spending over time. We refer to these as deflated estimates. However, for more detailed expenditure items this is not possible and we therefore compare unadjusted figures. Please see the Quality and methodology section for more information on comparisons over time.

Statistical significance testing indicates the probability we are confident that the difference between the estimates we are comparing did not occur by chance. Any change that has been described as “highly significant” is significant at the 99% level. This means that the probability that the difference occurred by chance is very low (1 in 100 or lower). Any change that has been described as “significant” is significant at the 95% level. This means that the probability that the difference occurred by chance is low (1 in 20 or lower).

This bulletin considers expenditure by UK constituent country and English region. We combine three years of data when presenting spending broken down by country and region to improve the robustness of results; these averages do not use deflated data.

This bulletin considers household expenditure by age group. The age of a household refers to the age of the person who is legally responsible for the household, known as the household reference person (HRP). Where there is more than one person who is legally responsible in the household, the HRP is the person with the highest income. If there is more than one person with the same income, the eldest member of the household becomes the HRP.

This article discusses median expenditure in addition to income decile and quintile groups. The median of a distribution divides it into two equal parts, so that half the households in a distribution of household expenditure will have higher expenditure than the median, and the other half will have lower expenditure than the median. Similarly, income quintiles and deciles divide the household income distribution into 5 and 10 equal parts respectively.

We use equivalised disposable income decile and quintile groups when looking at expenditure and income together. Disposable income is defined as gross weekly cash income less the statutory deductions and payments of Income Tax and National Insurance contributions. It is used alongside expenditure as it is the amount households have available to spend or save. Equivalisation takes into account that households with many members are likely to need a higher income to achieve the same standard of living as households with fewer members. Please see the Quality and methodology section for more information on equivalisation.

Family Spending has been designated by the UK Statistics Authority as National Statistics, in accordance with the Statistics and Registration Service Act 2007 and signifying compliance with the Code of Practice for Official Statistics.

Total average weekly household spending rose to £554.20 in the financial year ending (FYE) 2017. After adjusting for inflation, household spending has not been this high since the FYE 2006. It is also a rise of £21.20 in real terms when compared with the previous year. Figure 1 shows that household spending has exceeded pre-economic downturn levels for the first time.

By looking at headline economic indicators, we can put the latest household spending estimates into a wider context. Inflation, as measured by the Consumer Prices Index including owner occupiers’ housing costs (CPIH), grew from 0.7% in April 2016 to 2.3% in March 2017, the highest level of inflation seen since September 2013. The Consumer Prices Index (CPI) also measured 2.3% in March 2017. Inflation measures the change in prices of goods and services bought by households. Rising levels of inflation can encourage households to bring forward certain purchases; this in turn boosts economic growth. This could be one of the reasons behind higher levels of spending in FYE 2017.

Gross domestic product (GDP), a headline measure of how well the economy is performing, grew steadily for all quarters in FYE 2017. Quarter 1 (Jan to Mar) 2017 was the 17th consecutive quarter of growth, continuing the UK’s period of growth since Quarter 1 2013. Employment also continued to grow during this period. The employment rate reached 74.8% in Quarter 1 2017; the highest rate since records began at that time. The higher the number of people and households there are in work, the more households there are receiving a regular wage; this could be another factor that influenced higher levels of spending in this time period.

For FYE 2017, after taking account of inflation and changes in household structures over time, median disposable income increased by £600 (or 2.3%) when compared with the previous year. This shows that households had more money to spend when compared with the previous year.

Alongside these main economic indicators, there are other indicators we can consider to help build a wider picture of the economy. In 2016, the households’ savings ratio fell to 7%, the lowest seen since 2006. This shows that households chose not to save their money during this time period, which could be attributed to the low base interest rate of 0.25% introduced in August 2016. A low interest rate is a disincentive to save but it is an incentive to spend. In particular, it can encourage households to buy more goods and services on credit, as well as taking out loans, which would influence higher levels of household spending.

Household expenditure by category

After adjusting for inflation, the majority of estimates for classification of individual consumption by purpose (COICOP) categories increased in FYE 2017 when compared with FYE 2016. Expenditure on food and non-alcoholic drinks, transport, communication, and restaurants and hotels saw highly significant increases; whilst spending on clothing and footwear, household goods and services, and recreation and culture also saw significant increases. This suggests that the increase in spending at the overall level was driven by a general rise across nearly all goods and services. Figure 2 shows the deflated time series for each COICOP category.

Figure 2: Household expenditure at financial year ending 2017 prices, by classification of individual consumption by purpose category

UK, financial year ending 2002 to financial year ending 2017

Transport was the highest spending category in FYE 2017. On average, households spent £79.70 a week on this category. This was a highly significant increase of £5.40 when compared with the previous year after figures are adjusted for inflation. Transport includes the sub-categories:

• purchase of vehicles

• the cost of running a vehicle

• transport services (such as public transport and air fares)

All of these sub-categories increased when compared with the previous year (figures not adjusted for inflation), with the highest increase seen in the purchase of vehicles. Purchase of new vehicles expenditure increased by £1.80 when compared with the previous year, whilst spending on the purchase of second-hand vehicles increased by £3.10, figures not adjusted for inflation. As reported by the Society of Motor Manufacturers, 2016 was a record year for both new car purchases and second-hand car purchases alike. As shown in Table A45, the percentage of households that own a car or van has increased by 3 percentage points over the last three years to reach 79% in FYE 2017. Figure 3 shows how car and van ownership has increased since 1970.

We have seen an increase in spending on cars bought outright and cars bought on a hire loan purchase scheme, commonly referred to as Personal Contract Purchases (PCPs). The popularity of these schemes has also been observed by economists of the Bank of England. We see that expenditure on hire loan purchase agreements has particularly increased for second-hand cars, which increased from £4.40 on average per week to £6.50 when compared with the previous year (figures not adjusted for inflation).

Recreation and culture was the second highest spending category for the first time, where households spent an average of £73.50 a week on goods and services that fell into this category. This was a significant increase of £5.00 when compared with FYE 2016 estimates, after figures have been adjusted for inflation. The increase seen in FYE 2017 was largely driven by an increase in spending on package holidays abroad. Households spent an average of £25.50 a week on package holidays abroad, an increase of £4.70 when compared with FYE 2016 estimates (figures not adjusted for inflation). This is a percentage increase of 23%.

We saw evidence in other spending categories that households were spending more on trips abroad. For example, spending on accommodation services abroad rose by £1.60 and spending on international air fares increased by £0.40, where figures have not been adjusted for inflation. This could reflect the increasing number of visits abroad that were made by UK residents in 2016, as reported on in our Travel trends publication.

Many other goods and services fall into the recreation and culture category, where some saw a rise and some saw a fall when compared with the previous year. For example, average weekly household spending on pets and pet food rose by £0.80 to reach £5.40, whilst money spent on the lottery reduced by £0.30 to £1.50 (figures not adjusted for inflation).

Figure 4 is an interactive tree map that enables you to explore average weekly household spending for FYE 2017 in more detail.

Figure 4: Household expenditure in the UK, financial year ending 2017

When we refer to average weekly household spending in this bulletin, we use the mean to calculate the average. However, we can also consider median weekly household spending. Median household spending is the spending of the middle household, after we have sorted the households from those who spent the least to those who spent the most. Unlike the mean, the median is not influenced by a few households with very high spending.

For the FYE 2017, median weekly household spending was £461.30, an increase of £29.50 when compared with the median of the previous year (figures not adjusted for inflation). Median household spending was lower than the mean household spending for all COICOP categories. However, weekly household spending on communication, and food and non-alcoholic drinks had a similar value for both the median and the mean. This reflects that there were relatively few households spending a large amount of money on these categories.

Celebrating 60 years of Family Spending

The FYE 2017 edition of Family Spending marks the 60th year of results from a UK household expenditure survey. Since the survey started in 1957, approximately 375,000 households have provided information on their spending habits – thank you to all who have taken part.

The first household expenditure survey was an extension of the National Food Survey (NFS), which recently celebrated its 75th birthday. Introducing a household expenditure survey that collected information on all types of household expenditure enabled the government to improve the measure of inflation and also provided expenditure information in relation to income and family circumstances. Through the years, the survey has gone through various changes, but we are still able to look at some trends over the 60-year time period. Unlike the rest of this bulletin, expenditure is classified by the original Family Expenditure Survey (FES) classifications (not COICOP) to compare over the 60 years. Please see the Quality and methodology section for information on the methodology used for our 60-year time series.

We can see that when using the Family Expenditure Survey (FES) classification of housing (this includes net rent, mortgage interest payments, Council Tax and repairs, maintenance and decoration), the proportion of total expenditure on housing has doubled during the last 60 years from 9% to 18%. (Please note that prior to 1992, expenditure for owner-occupiers and those who lived in rent-free accommodation was imputed.)

On the other hand, the proportion of total expenditure on food has halved (33% to 16%), as has the proportion of total expenditure on clothing (10% to 5%). This may be due to the increased choice of readily available goods and services, offered at a wider range of prices. In addition to this, the census shows that the average household size has decreased over the 60 years, which would also influence average weekly household spending on food and clothing.

In 1957, average weekly household expenditure on tobacco made up 6% of total spending. For FYE 2017 this reduced to 1%, which mirrors the downward trend seen in the number of people who smoke in Great Britain. Spending on alcohol has risen and then declined over the 60-year period. In 1957, the proportion of total expenditure on alcohol was 3%, before rising to 5% in the 1970s and 1980s. This then declined, with the most recent data showing that the proportion of total expenditure on alcohol was 3%; the same as 1957.

The estimates in the first section of this bulletin have concentrated on looking at the average weekly spending of households in the UK as a whole. The rest of this bulletin will consider the spending patterns of different types of households to provide a clearer picture of the varied spending priorities, pressures and choices of these different groups.

This section will consider how household spending varies between the UK constituent countries, that is, England, Wales, Scotland and Northern Ireland. When looking at spending by country, we combine three years of financial year data to make the analysis more robust; these averages do not use deflated data.

The different countries of the UK have different spending patterns when compared with one another. This is likely to be the result of differences in income, differences in lifestyle, and regional price differences.

Table A33 shows that England had the highest level of household spending for the financial year ending (FYE) 2015 to FYE 2017. Average weekly household spending for England was £547.40 a week, compared with the UK average of £536.80 for the same time period. England was the only UK constituent country whose average weekly household spending was more than £500, where five out of the nine English regions spent more than £500 a week.

However, spending within England varied greatly. Average weekly household spending in London and the South East was over £600 whilst average weekly household spending in the North East was approximately £200 less at £437.00. The most recent population estimates show that the English population is currently 55.2 million people; this is 84% of the UK population. It is therefore not surprising that there are large differences within the country.

Figure 5 demonstrates the differences in overall spending for the different English regions and compares them to Wales, Scotland and Northern Ireland. This shows that when we consider the UK as a whole, England had some of the lowest and highest spending regions.

The top three spending categories for England were transport, housing, fuel and power, and recreation and culture. Unlike the UK as a whole, English households spent more on housing, fuel and power than recreation and culture. This was driven by higher spending on rent in London. As shown by Table 2.11, renters in London spent an average weekly amount of £189.00 on net rent, where net rent is the amount payable by a household after housing benefits have been deducted. This averages out rent across all renting households, not all households. This was more than two and a half times the average amount spent by renters in Wales, Scotland and Northern Ireland.

However, not all regions of England spent less on recreation and culture when compared with other categories. In fact, recreation and culture was the top spending category for the North East, the North West and Yorkshire and The Humber. These three regions were the only regions of the UK that had recreation and culture as the top spending category. This was driven by a few different factors, such as lower spending on housing and a higher proportion of total expenditure on package holidays.

Following on from England, the second highest spending UK constituent country was Northern Ireland. Northern Ireland spent an average of £497.10 a week in FYE 2015 to FYE 2017. Northern Ireland was the only country of the UK that had food and non-alcoholic drinks in its top three spending categories.

Households in Northern Ireland spent 13% of their total expenditure on food and non-alcoholic drinks, compared with the UK average of 11%. One reason for this is that the average Northern Ireland household size is larger than the UK average. Cultural differences may also be a factor here, where households in Northern Ireland consider items such as good quality cuts of meat to be more of an essential food item compared with those in England. Northern Ireland households also spent a higher proportion of total expenditure on restaurant and café meals and take-away meals when compared with England, Wales and Scotland.

Scottish households spent an average of £492.30 a week in FYE 2015 to FYE 2017; this is similar to the amount spent by households in Northern Ireland. Households in Scotland spent 15% of total expenditure on transport, making this the top Scottish spending category. Scottish households spent 2.5% of their total spending on buying a new car or van, which is 0.8 percentage points higher than the UK average. However, they spent less of their total expenditure on second-hand cars when compared with other areas of the UK.

Higher levels of transport spending in Scotland could be attributed to the growing rural population in Scotland. As shown by Table A36, households in rural areas spent a greater proportion of total expenditure on transport. Fewer people in Scottish rural areas were satisfied with public transport than the rest of Scotland in 2016, which could be one reason behind the higher spending on new cars. Although transport was the top spending category in Scotland, Scottish household average weekly spending on recreation and culture was only £1.30 less. This was particularly driven by Scottish households spending £26.20 on package holidays, which is an increase of £5.40 when compared with the previous year (figures not adjusted for inflation).

Households in Wales had the lowest average weekly spending out of the four UK constituent countries for FYE 2015, FYE 2016 and FYE 2017. Here, average weekly household spending was £458.70; this was £78.10 less than the UK average for this three-year time period. When we compare Wales’ average weekly household spending with the separate regions of England in addition to Scotland and Northern Ireland, the North East was the only region that spent less than households in Wales, at £437.00 a week.

The top spending category for Wales was transport, whilst recreation and culture, and housing, fuel and power ranked joint second. This mirrors the high car ownership we see in Wales; Table A48 shows that 83% of households in Wales owned a car or van in this three-year period. This was higher than England (77%), Scotland (74%) and Northern Ireland (79%). Households in Wales spent an average of £22.90 a week on the purchase of vehicles, which is 5% of total expenditure. The majority of this was spent on second-hand cars and vans, which is different to the pattern we have seen in Scotland.

In addition to looking at household spending by area, we can look at how household spending varies by age group. The rest of this bulletin considers the spending habits of some of the oldest members of our society, as well as some of the youngest.

The latest UK population estimates show that the UK population is ageing; 18% of the population are now aged 65 and over. This section will concentrate on the spending of households where the person legally responsible for the household, the household reference person (HRP), is aged 65 to 74 years or aged 75 and over.

As age increases, spending pressures and priorities of households change. For example, as shown in Table 2.4, the proportion of total expenditure on housing costs decreases with age. The housing costs in this table include goods and services such as rent and mortgage payments, alterations to dwellings and household insurance. This is a departure from the classification of individual consumption by purpose (COICOP) definition to provide a more complete picture of housing costs. In the financial year ending (FYE) 2017, whilst 27% of total expenditure was spent on housing costs for households in the under 30 years age bracket, this reduced to 19% for households aged 65 to 74 years and 18% for households aged 75 and over. This reflects that a greater percentage of older households own their own home outright.

Table A10 shows that households in the oldest age bracket (where the HRP was aged 75 and over) spent 14% of total spending on food and non-alcoholic drink in FYE 2017. This is a much larger proportion of total spending when we compare with the younger age groups, as demonstrated by Figure 6.

As shown by Table A16DE, households aged 75 and over in the lowest equivalised income group spent 17% of their total expenditure on food and non-alcoholic drink in FYE 2017, compared with 9% in the highest equivalised income group. This demonstrates how spending pressures and priorities can vary depending on income regardless of age group.

Table A10 shows that households in the aged 75 and over category also spent 14% of total expenditure on recreation and culture in FYE 2017. This is a high proportion of total spending when compared with households aged less than 65 years. In fact, households with the lowest income in this oldest age bracket spent the same proportion of total expenditure on recreation and culture (10%) as those in the highest income group of the youngest age category (also 10%). When compared with younger age groups, households aged 75 and over spent proportionally more of their total expenditure on goods and services such as package holidays in the UK and newspapers, books and stationery.

In the FYE 2017, the only age group that spent proportionally more on recreation and culture than households in the aged 75 and over category was the 65-to-74-year-olds.

The 65-to-74-year-old households spent nearly a fifth (18%) of total spending on recreation and culture, making this the top expenditure category for this age group. When compared with the other age groups, Table A11 shows that they spent the largest proportion of total expenditure on package holidays abroad and pets and pet food. Households in the 65-to-74 years age bracket may have more time to spend on recreation and culture goods and services when compared with the other age groups, as they are more likely to be retired and still active.

Latest results from the UK Harmonised European Time Survey show that the amount of time spent on leisure peaked for those aged 65 and over, where this age group took an average of seven hours and 10 minutes leisure time per day. Many grandparents also provide informal care for their grandchildren, which could explain the higher spending on recreation goods and services.

One effect of an ageing population is the increase in State Pension age. We are able to look at the difference in household spending between retired households that are mainly dependent on State Pension and retired households that are not. The latest household disposable income and inequality statistics show that there has been an increase in inequality for retired households in recent years, due in part to a growing gap between households that have private pensions and those that do not.

Table A23 shows that the average weekly household spending of retired households not mainly dependent on a State Pension was 1.4 times the amount spent by retired households that are mainly dependent on State Pension. Those mainly dependent on State Pension spent a greater proportion of their total expenditure on food and non-alcoholic drinks. This was particularly the case for one-person households; these households spent 16% of their total spending on food and non-alcoholic drinks, compared with 12% for one-person households not mainly dependent on a State Pension.

Households mainly dependent on State Pension also spent less of their total expenditure on recreation and culture. Interestingly, this difference was not so apparent for spending on restaurants and hotels, where the proportion of total spending was more equal between the two different types of retired households.

Households in the older age brackets have different spending pressures and priorities when compared with other age groups. To understand the other age groups in more detail, the next section of this bulletin compares the spending patterns of households with and without children.

Approximately 42% of all UK families contain dependent children, where a family is defined to be a couple with or without children, or a lone parent with at least one child, who live at the same address. This section of the bulletin looks at how household spending varied for households with and without children in the financial year ending (FYE) 2017. For the purposes of this section, a child is defined to be any household member under the age of 18 years.

We will firstly look at one-adult households with and without children. The number of lone parent families in the UK in 2016 was estimated to be 2.8 million, where 55% of these families contained only one child. Table A23 shows that in FYE 2017, one-adult households with one child spent an average of £10.80 less a week than one-adult non-retired households with no children. There is evidence to suggest that one-person households without children spent more on discretionary items. For example, one-adult households without children spent, on average, £5.10 more a week on recreation and culture when compared with one-adult households with one child. They also spent £4.90 more a week on restaurants and hotels. These households also spent £8.90 less on food and non-alcoholic drinks when compared with one-adult households with one child. This suggests that one-adult households without children are able to spend more on discretionary items as they don’t have to spend as much on essential items such as food.

However, when we consider one-adult households with two or more children, we see a slightly different and more complex picture. This type of household spent an average weekly amount of £61.00 on recreation and culture. This was 15% of total expenditure, whilst one-person households without children spent a much smaller 11%. Here, we see that one-adult households with two or more children were spending a greater proportion of their total spending on certain discretionary items.

When we consider the spending on different classification of individual consumption by purpose (COICOP) categories by households with and without children, transport had one of the biggest differences. Figure 8 shows how transport spending in FYE 2017 was different for households with and without children.

One-adult households without children spent 13% of total expenditure on transport compared with 11% for one-adult, one-child households and 10% for one adult, two or more children households in FYE 2017. This trend was also seen between two-adult households with and without children. Two-adult households without children spent 16% of their total expenditure on transport, compared with 14% for two-adult households with children. This reflects that adult households without children drove greater distances than those with children, as well as travelling further distances on public transport. In addition to this, households with children travelled more often on foot.

Perhaps unsurprisingly, households with children spent a greater proportion of their total spending on clothing and footwear. However, when we consider this in more detail, we see that this is only true for households in lower equivalised income quintiles. Tables 3.15E and 3.16E show that households with children in the lowest equivalised income quintile spent approximately 6% of total expenditure on clothing and footwear, whilst households without children spent 3%.

In contrast with this, both households with and without children in the top equivalised income quintiles spent 4% of total expenditure on clothing and footwear. This reflects how certain items can be viewed as both essential and discretionary items. Whilst all households would need to spend some money on clothing and footwear, households with a larger disposable income are able to choose to spend more on more expensive clothing and footwear items, or perhaps buy a greater amount.

Understanding the spending habits of UK children

As well as looking at expenditure of households with and without children, we can look at the spending patterns of children themselves. Unlike the rest of this bulletin, this section will concentrate on individual (not household) spending patterns, concentrating on those aged 7 to 15 years.

Children aged 7 to 15 years were asked to keep a diary of everything they spent their money on within a two-week period. This could be money they received through pocket money, gifts or through a job such as a paper round. In addition to reporting on average weekly child spending, this section looks at the most frequently bought items within the diary keeping two-week period. To improve the robustness of results, we have joined together three years of financial year data where figures have not been adjusted for inflation (financial year ending (FYE) 2015, FYE 2016 and FYE 2017). Please see the Quality and methodology section for more information on the methodology used in this section.

On average, children aged 7 to 15 years spent £12.40 a week for FYE 2015, FYE 2016 and FYE 2017. When we break spending down by single year of age, we can see that spending increased with age, as shown by Figure 9.

Children aged 15 years spent an average of £25.00 a week, which is more than three times the average spending of a 7-year-old.

We can break this down further and consider average weekly child spending by grouped age and sex as shown by Figure 10.

Figure 10 shows that there is no clear pattern when we consider average weekly spending between the sexes for the first two age groups. In the youngest age group, boys spent an average of £1.00 more than girls, whilst in the next age group the story is reversed where girls spent an average of £0.75 more than boys. However, when children enter the teenage years the gap begins to widen, where girls spent an average of £2.80 a week more than boys.

One type of good that stands out when we compare average spending by sex and age is toiletries and cosmetics. This includes things such as soap, shampoo and makeup. Average weekly spending on toiletries and cosmetics was less than £0.10 for boys in all of the grouped age categories. In contrast with this, the average weekly spending on toiletries and cosmetics rose from £0.20 for 7-to-9-year-old girls, to £0.50 in the next age group and finally to £1.70 for girls aged 13 to 15 years. Considering this further, we see that girls bought toiletries and cosmetics more frequently than boys. Only 2% of 7-to-15-year-old boys bought at least one toiletry or cosmetic item in a two-week period, compared with 17% of girls of the same age. This was true for all subcomponents of the larger toiletry and cosmetic category. For example, 13% of 7-to-15-year-old girls bought at least one cosmetic or related accessory item in a two-week period, compared with 1% of boys of the same age.

Other items that had notable differences between boys and girls were books and computer games and software. Average weekly spending on books by girls aged 7 to 15 years was nearly double that of boys (£0.30 compared with £0.15 respectively), whilst boys spent more than 10 times the amount on computer games and software when compared with girls (£1.10 compared with £0.10 respectively).

One of the most frequently purchased items by children aged 7 to 15 years was soft drinks, which includes still and fizzy drinks such as fruit juice or lemonade. Over 56% of children aged between 7 and 15 years bought at least one soft drink within a two-week period, with the majority of drinks being consumed away from the home. The average weekly child spending on soft drinks was £0.90. We see that the popularity of buying soft drinks increased with age; 75% of 13-to-15-year-olds bought at least one soft drink within a fortnight, compared with 58% of 10-to-12-year-olds and 38% of 7-to-9-year-olds.

Although the younger age group bought soft drinks less frequently, this was not true for all items. The younger age group bought items that fell into the games, toys and hobbies category more frequently than the other age groups. This includes items such as jigsaw puzzles, teddy bears and card games. We see that more than 27% of children aged 7 to 9 years bought at least one item that fell into this category within a fortnight, compared with 14% of 10-to-12-year-olds and 5% of 13-to-15-year-olds. We also see that ice cream was more popular with the younger age group, where 22% bought at least one ice cream, compared with 15% of 13-to-15-year-olds.

As well as low-cost items such as soft drinks, sweets and games, we see that children bought more expensive items. Some of the more common expensive items bought by children in this time period included bikes, mobile phones and personal computers.

Details of household consumption expenditure within the context of the UK National Accounts are produced as part of Consumer trends. This publication includes all expenditure by members of UK- resident households. National accounts figures draw on a number of sources including the Living Costs and Food Survey (LCF): figures shown in this report are therefore not directly comparable with national accounts data. National accounts data may be more appropriate for deriving long-term trends on expenditure.

Further information about food consumption and in particular details of food quantities, is available from the Department for Environment, Food and Rural Affairs, who produce their own report of the survey.

The Household income and inequality and Effects of taxes and benefits on household income bulletins provide a more detailed analysis of disposable income and inequality in the UK. Our preferred measures of household income are published in these releases. Both these releases use data provided by the LCF, the same data source used in Family Spending. The income measures follow the definitions and concepts set out in the Canberra Handbook (published by the United Nations Economic Commission for Europe (UNECE)), which is the basis of internationally agreed standards in this area. This defines income as receipts (either monetary or in kind) that are received on a regular basis and are available for current consumption.

The financial year ending March 2017 edition of Family Spending includes some changes to the range and content of available tables; this is to ensure robust estimates. An index of tables gives more information.

Slight changes have been made to the way we deflate our time series. Please see the Quality and methodology section for more information.

The Living Costs and Food Survey Technical Report and corresponding Quality and Methodology Information report, contains important information on:

• the strengths and limitations of the data and how it compares with related data

• uses and users of the data

• how the output was created

• the quality of the output including the accuracy of the data

The financial year ending (FYE) 2017 version of the technical report will be updated shortly.

Description of the survey

All the findings in this bulletin are taken from data collected on the Living Costs and Food Survey (LCF). The LCF is a UK household survey, designed to provide information on household expenditure patterns and food consumption.

The LCF is a voluntary sample survey of private households. Each individual in a selected household is asked to complete a household interview and then an expenditure diary for two weeks. The survey is continuous, interviews being spread evenly over the year to ensure that seasonal effects are covered.

Great care is taken to ensure complete confidentiality of information and to protect the identity of LCF households. Only anonymised data are supplied to users.

Sampling

The LCF sample for Great Britain is a multi-stage stratified random sample with clustering. It is drawn from the small users file of the postcode address file (PAF) – the Post Office’s list of addresses. After stratification by two census variables, 639 postal sectors are randomly selected and, from within these, households are sampled.

A total of 11,505 households were selected in FYE 2017 for the LCF in Great Britain.

The Northern Ireland sample is drawn as a random sample of addresses from the Land and Property Services Agency list.

Response: Great Britain

The overall response rate for the LCF in FYE 2017 was 45% in Great Britain.

Because of the practicalities of running a voluntary survey, it is not possible to get full response from every selected household. Of the total 11,505 households in Great Britain selected for the LCF in FYE 2017, there were 4,641 households that co-operated fully in the survey, including 125 households that provided sufficient information to be included as valid responses.

Response: Northern Ireland

The overall response rate for the LCF in FYE 2017 was 45% in Northern Ireland.

In FYE 2017, there were 1,000 sampled households, 889 of which were eligible, where 389 households gave full interviews.

Estimation

The LCF employs weighting to reduce the effect of non-response bias and produce estimates of population totals and means. The weighting procedure is designed to compensate for non-response (sample-based weighting) and to adjust the sample to reflect known population totals (population-based weighting).

The technical report goes into further detail showing the effects of the weighting by comparing unweighted and weighted data.

Standard errors and estimates of precision

The LCF is a sample of households and not a census of the whole population. Therefore, the estimates presented in this report will differ, to some degree, from the true population values. For more information on all estimates of precision, please see the technical report.

Standard errors provide a numerical measure of the likely difference that will arise, solely due to random chance. Standard errors for detailed expenditure items are presented in relative terms in Table A1 (standard error as a percentage of the average to which it refers). Please note that when broken down further by sub-category, for example, household composition, the standard errors will increase; therefore, extra care should be taken when interpreting and comparing figures.

Confidence intervals can be used to estimate a range within which the true population value should lie. This report presents confidence intervals at the 95% level. This means that if we used the same sampling method to select different samples and computed an interval estimate for each sample, we would expect the true population parameter to fall within the interval estimates 95% of the time.

Statistical significance testing indicates the probability with which we are confident that the difference between the estimates under examination did not occur by chance. Significant testing is carried out at the 95% and 99% levels. Please see the Things you need to know about this release section for definitions of highly significant and significant results.

Equivalisation methodology

Income quintile and decile groups are based on a ranking of households by equivalised disposable income. Equivalisation is the process of accounting for the fact that households with many members are likely to need a higher income to achieve the same standard of living as households with fewer members. Equivalisation takes into account the number of people living in the household and their ages, acknowledging that while a household with two people in it will need more money to sustain the same living standards as one with a single person, the two-person household is unlikely to need double the income.

The Family Spending Technical Report provides further detail.

Spending by UK constituent country and English region

This bulletin considers expenditure by UK constituent country and English region. We combine three years of data when presenting spending broken down by country and region. This is done to increase the sample size and therefore improve the robustness of results. The average for each year is summed and then the total is divided by three. The averages do not use deflated data.

Comparisons over time

Throughout this bulletin, figures are compared over time. Where possible, expenditure figures have been deflated, that is, adjusted for inflation, to allow comparison of expenditure in real terms across survey years. In this bulletin, all figures have been deflated when comparing the main categories of the classification of individual consumption by purpose (COICOP) over time. Figures not adjusted for inflation are also available for the main COICOP categories in Table 4.3. More detailed categories have not been deflated; this is stated in the commentary. Figures are deflated using the Consumer Prices Index (CPI) and the indices are specific to each major COICOP category.

For housing goods and services items that are not covered by the CPI, the owner occupier housing costs (payments method) indices are used. Please note that this is new for this release; in previous years categories not covered by CPI used the Retail Price Index. The LCF technical report contains greater detail on how our figures are deflated. Please note that other ONS statistics referenced in the commentary may use a different method for deflating time series figures, according to the concept being measured.

The LCF is reviewed every year and changes are made to keep it up-to-date. Therefore, year-on-year changes should be interpreted with caution.

Standard errors for categories with lower levels of spending tend to be higher, so trends for these categories need to be treated with a degree of caution.

60-year time series

This bulletin looks at how spending has changed since 1957. For this section of the bulletin, the raw spending estimates from 1957 onwards were taken and converted to pounds sterling where applicable. These estimates were then deflated using the RPI All Item Index, based to FYE 2017 prices.

Please note that over the years, classifications of items and groups of items have changed. The grouping used in the 60-year section of this bulletin uses the original Family Expenditure Survey (FES) classifications to compare over time. In 1986, FES groupings changed to align with the categories recommended for the RPI where the 11 categories were extended to 14. The financial year ending 2002 Volume 5 user guide contains information on how the FES classifications map to the COICOP classifications, the classification used for the majority of this bulletin.

Child spending estimates

This bulletin contains estimates for child spending. These estimates were created using an individual weight created for this analysis. Each responding child was allocated the design weight of the household, which was then adjusted for non-response. This was then calibrated to the 7-to-15-year-old population totals. More information on how this weight was created can be found in the technical report.

Any money given by an adult to a diary-keeping child in the same household is subtracted from household expenditure totals to avoid double counting.

We combine three years of data when presenting child spending. This is done to increase the sample size and therefore improve the robustness of results. The average for each year is summed and then the total is divided by three, where the averages do not use deflated data.