Stats NZ has a new website.

For new releases go to

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
A comparison of household spending patterns based on the 1980 and 2008 CPI baskets

Household spending habits change over time. New technology delivers a seemingly endless supply of new consumer goods to the market, while those that fall out of favour disappear. The range of goods and services available for household consumption is continuously changing and this, in combination with changing consumer preferences and tastes, results in changes to household spending patterns. This article examines how consumer spending habits have changed over the past three decades by looking at the differences between the 1980 and 2008 consumers price index (CPI) baskets.

The CPI basket refers to the goods and services that have been selected to represent all of the goods and services that households buy. The CPI basket is updated periodically to ensure it accurately reflects household spending patterns. In addition to updating the basket, the weights assigned to individual goods and services are updated, so that, for instance, if there was an increase in the relative amount households spend on petrol compared with newspapers then the weight for petrol would increase relative to newspapers. The basket and weights can be used to examine how household spending habits have changed over time.

Since the June 2006 quarter the CPI basket has been classified using the New Zealand Household Expenditure Classification (NZHEC). This classification assigns each good and service purchased by households into one of the following groups:

  • food
  • alcoholic beverages and tobacco
  • clothing and footwear
  • housing and household utilities
  • household contents and services
  • health
  • transport
  • communication
  • recreation and culture
  • education
  • miscellaneous goods and services.

Before 2006 another CPI commodity classification was used, therefore in order to compare the 1980 and 2008 baskets the basket items from 1980 have been reclassified using the NZHEC. Mortgage interest payments and sections were both included in the 1980 basket but are now out of scope of the CPI, so these have been excluded from the analysis and the weights of the remainder of the basket has been scaled accordingly.

Table 1 shows the percent of total household expenditure each NZHEC group accounted for in both 1980 and 2008. It illustrates the proportion of total spending that households spent on each group. It is important to note that while the overall expenditure amount has increased significantly over this time, it is the proportions of each group to total that are comparable and of interest.

Table 1 

 Consumers Price Index
 Expenditure weights 1980 and 2008
 Group 1980  2008 Percentage points change 
 Food 19.70 17.83 -1.87 
 Alcoholic beverages and tobacco  9.13  6.76 -2.36
 Clothing and footwear  7.30 4.48 -2.83
 Housing and household utilities 16.63 22.75 6.12
 Household conents and services  9.17 5.26 -3.91
 Health  1.46 5.09 3.62
 Transport  16.93 16.18 -0.75
 Communication 1.46 3.21 1.75
 Recreation and culture 8.04 9.54 1.50
 Education 0.36 1.78 1.42
 Miscellaneous goods and services 9.80 7.12 -2.68
 All groups 100.00 100.00 0.00

Note: Figures may not sum to totals due to rounding.

The transport group has a relatively small difference in its proportions between 1980 and 2008. Households are still spending a similar proportion of their total spending on goods and services, within the scope of the current CPI, on transport now as they were in 1980.

Others groups, for instance, clothing and footwear, and household contents and services have shown significant falls in the proportion of total spending. Households are now spending proportionally less on these groups, compared with other groups, than they were in 1980. These falls coincide with reductions in import tariffs, in particular for clothing and footwear, which have been falling since the 1980s. In 1987 the import tariff for clothing was set at 65 percent for some clothing items, and 40 percent for others. By 1991 the tariff had fallen to 40 percent for all items and by 1999 it was down to 19 percent. Likewise, footwear tariffs fell from 55 percent in 1991 to 19 percent in 1999(1).

Groups such as communication and education now have proportionally higher spending on them than previously. Households are now spending a higher proportion on these groups than they were in 1980.

The increased proportion of spending on the education group is consistent with an increase in both the volume and price of tertiary education. New Zealand's official year-book shows that tertiary education numbers have increased significantly since the early 1980s. In the decade between 1980 and 1990 the number of students attending tertiary education increased by 55 percent, followed by a further 154 percent increase between 1990 and 1999(2). Ministry of Education data shows a further increase in the number of enrolled equivalent full-time students of 32 percent between 1999 and 2008(3). In addition, the tertiary and other post-school education class of the CPI increased by 185.9 percent between the December 1993 quarter (when the series started) and the March 2008 quarter. Changes to the volumes and prices of primary and secondary education have not been as significant as tertiary education.

For the health group it is important to note that in 2002 there was a conceptual change in the way insurance was dealt with in the CPI. From 2002 onwards, net weights for insurance were introduced, whereby the weight of insurance is measured by the service charge households pay to be insured. Under the net approach any expenditure made by insurance companies on behalf of households (claims made by households) is allocated directly to the weight of that particular good or service. This had the effect of increasing the weights on the health group, in particular out-patient and hospital services. The health group weight went from 2.20 percent of the CPI at the 1999 review, to 4.83 percent at the 2002 review.

Changes to the CPI basket can also be used to compare how spending within each of these groups has changed. Table 2 summarises the changes within each group between 1980 and 2008.

Table 2

   Consumers Price Index
       Expenditure weights 1980 and 2008
    Groups and subgroup
Group / Subgroup Percent of group weight
1980 2008
Food 100 100
Fruit and vegetables 14 13
Meat, poultry and fish 24 17
Grocery food 38 39
Non-alcoholic beverages 7 10
Restaurant meals and ready-to-eat food 18 21
Alcoholic beverages and tobacco 100 100
Alcoholic beverages 74 68
Cigarettes and tobacco 26 32
Clothing and footwear 100 100
Clothing 79 82
Footwear 21 18
Housing and household utilities 100 100
Actual rentals for housing 19 35
Home ownership 32 24
Property maintenance 24 12
Property rates and related services 10 12
Household energy 15 18
Household contents and services 100 100
Furniture, furnishings and floor coverings 31 37
Household textiles 9 8
Household appliances 22 22
Glassware, tableware and household utensils 10 6
Tools and equipment for house and garden 11 10
Other household supplies and services 17 17
Health 100 100
Medical products, appliances and equipment 22 19
Out-patient services 72 65
Hospital services 6 15
Transport 100 100
Purchase of vehicles 31 25
Private transport supplies and services 52 57
Passenger transport services 18 18
Communication 100 100
Postal services 16 5
Telecommunication equipment 0 6
Telecommunication services 84 88
Recreation and culture 100 100
Audio-visual and computing equipment 22 19
Major recreational and cultural equipment 11 4
Other recreational equipment and supplies 24 21
Recreational and cultural services 16 25
Newspapers, books and stationery 18 14
Accommodation services 7 7
Package holidays 1 10
Education 100 100
Primary, secondary and tertiary tuition and examination fees 100 ..
Early childhood education .. 14
Primary and secondary education .. 37
Tertiary and other post-school education .. 47
Other education .. 2
Miscellaneous goods and services 100 100
Personal care 19 33
Personal effects 10 7
Insurance 19 26
Credit services 10 9
Other miscellaneous services 42 27

.. Not available
Note: Figures may not sum to totals due to rounding.

As was the case at the higher level, some of the subgroups have recorded little change in their relative proportions while others have shown large changes.

The food group has shown a shift away from fresh foods. In 1980, the combined weight of the fruit and vegetables and the meat, poultry and fish subgroups accounted for 37 percent of the total household spending on food. By 2008 this was 30 percent, a fall of 7 percentage points. In comparison, the weight for non-alcoholic beverages has increased from 7 percent in 1980 to 10 percent in 2008 and the weight of restaurant meals and ready-to-eat food increased from 18 to 21 percent. This illustrates that households are now spending proportionally less on fresh foods and more on non-alcoholic beverages, takeaway food and eating out than they were in 1980.

Another group to show a noticeable shift in the proportions of each subgroup is the housing and household utilities group. In 1980, housing rentals accounted for 19 percent of total household spending in this group, but by 2008 this had increased to 35 percent. In comparison, the proportion of household spending on home ownership has fallen from 32 percent in 1980 to 24 percent in 2008. This partly reflects a fall in the rate of home ownership. In the past 20 years, home ownership was highest in the 1991 Census, at 73.8 percent. By the 2006 Census, it had fallen to 66.9 percent. Note that home ownership in the CPI covers only the net additions to the housing stock. It does not include the purchase and sale of existing houses between private households, the effect of which, under the current CPI conceptual approach, is assumed to net out. The methodology used to capture net additions to the housing stock changed in 2006 to include an estimate of net additions from buying or selling owner-occupied dwellings to or from landlords, small businesses, developers, and government. Previously, net additions to the housing stock included only new residentual buildings. This resulted in a lower weight in 2006 for home ownership than would have been the case if the previous methodology had been used. The proportion of household spending on property maintenance, which includes both materials and services, has fallen from 24 percent in 1980 to just under 12 percent in 2008.

The heath group has shown some change in its proportions between 1980 and 2008. The largest change in this group is that a larger proportion of household spending on health is now on hospital services. Both medical, appliances and equipment, and out-patient services have shown small falls in their relative proportions. These changes are in part due to the 2002 change in the way expenditure by health insurance companies is dealt with, as explained earlier. This resulted in the weight for out-patient services increasing from 55 percent of the total health group weight in 1999 to 73 percent in 2002, and an increase from 4 percent to 14 percent for hospital services. The out-patient services weight fell to 65 percent of the total health group weight in the 2006 review, with a corresponding increase in the weight of medical products, appliances and equipment.

Transport is another group that has shown a shift in how households spend their money. While the proportion of spending on passenger transport services, which includes services such as public transport, taxi fares, and airfares, has stayed the same, the proportion of total transport spending on the purchase of vehicles has fallen from 31 percent in 1980 to 25 percent in 2008. There has been a corresponding increase in private transport supplies and services. The increase in the proportion of weight for private transport supplies and services is largely due to petrol and private transport services. Petrol prices have increased more strongly than the rest of the transport group between 1980 and 2008. In the March 1980 quarter the average price for a litre of 96 octane petrol was 46.3 cents. Taking into account transport inflation this was equal to $1.53 in the June 2008 quarter. The average price in the June 2008 quarter for a litre of 95/98 octane petrol was $2.00, giving an increase in the real price of petrol, relative to all transport goods and services, of 30.9 percent.

Similarly, the recreation and culture group has shown some change in the proportions spent on each of its subgroups. The biggest change is an increase in the proportion of total household spending on recreation and culture that is on package holidays, from around 1 percent in 1980 to around 10 percent in 2008. The majority of this shift occurred in 2006 when international package holidays prepaid in New Zealand were added to the basket. Previously, expenditure on these had been allocated to international airfares in the transport group. This change would have had a downward impact on the weight of the transport group. The weight of major recreational and sporting equipment has fallen from about 11 percent to about 4 percent.  All items within this subgroup have a smaller proportion than previously, with boats and outboard motors accounting for 2.8 percent of the total recreation and culture weight in 2008, compared with 5.1 percent for boats and boating equipment in 1980. Similarly, musical instruments accounted for 1.7 percent in 1980, but in 2008 this had fallen to 0.7 percent. Caravans, which have now been removed from the CPI basket, accounted for 4.1 percent of the recreation and culture weight in 1980. It is difficult to calculate accurate weights for this area of the basket as there are relatively few of these items purchased and therefore reported by respondents in the Household Economic Survey, the main source of data for the calculation of CPI weights. 

The miscellaneous goods and services group has also experienced a reasonable relative shift in spending between its subgroups. Household spending on personal care, which includes appliances and services for personal care, has increased from 19 percent of total miscellaneous goods and services spending to 33 percent. This is a result of increased expenditure on hairdressing and personal grooming services, and on non-electrical appliances for personal care. In the March 1980 quarter, the average price for a woman's haircut was $3.96. Taking inflation into account, this was equivalent to $17.03 in the June 2008 quarter. However, the June 2008 quarter average price for a woman's hair cut was $54.64, an increase of 221 percent in real terms. Men's haircuts have also shown a large increase in price, with a man's dry haircut increasing by 91 percent in real terms over the same period.

Also within the miscellaneous goods and services group spending on insurance increased from 19 percent in 1980 to 25 percent in 2008. Life insurance has been added to the basket since 1980, and now has the largest weight of all insurance items. The relative proportion of health insurance has increased slightly, while the proportions of dwelling, contents and vehicle insurance have all shown falls. The method for calculating general insurance (dwelling, contents and vehicle) weights was changed to the net approach in 2002, at the same time as the change in health insurance. This approach, which was further refined in 2006, had a downward impact on the weights of these insurance items.

While each of these groups has shown some change in its composition there are some that have had very little change between 1980 and 2008. One example of this is the clothing and footwear group. Clothing currently accounts for 82 percent of total spending in this group, which is very similar to the 79 percent it accounted for in 1980.

Similarly, the household contents and services group has shown very little change between the two years. In 2008 the household textiles, household appliances, tools and equipment for house and garden, and other household supplies and services subgroups all accounted for very similar proportions as what they did in 1980. However, the remaining two subgroups, furniture, furnishing and floor coverings (up from 31 percent to 37 percent), and glassware, tableware and household utensils (down from 10 percent to 6 percent) have both changed.

An additional subgroup has been introduced to the communication group since 1980 making comparison difficult. The telecommunication equipment subgroup, which accounted for 6 percent of the total group weight in 2008, was introduced in 1999 as a result of the introduction of the purchase of telephones to the CPI basket. In earlier decades, households tended to rent their home phones.

The historical data used for this analysis does not give a detailed breakdown of each item in the basket in 1980. While this has not affected the analysis of the majority of the basket it has affected the education group. Consequently, a comparison between the subgroups of the education group in 1980 and 2008 is not available.

By comparing the basket items and weights in both 1980 and 2008 a comparison can be made illustrating how household spending habits have changed over this time. Not only has household spending changed between the CPI groups, with households spending proportionally more on communication and education and less on clothing and footwear and household contents and services than they were in 1980, but it has also changed within the groups. This is particularly evident in the housing and household utilities group where falls in home ownership rates have resulted in a smaller weight for home ownership and a corresponding larger weight for housing rentals.

  1. Ministry of Economic Development, Impacts of previous tariff reduction [Online]
  2. Statistics New Zealand, Long-term data series [Online]
  3. Education Counts, Participation [Online]


Back to Price Index News: October 2009

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+