Series
Food Price Index
en-NZFPI
en-NZStats NZ
en-NZThe food price index (FPI) measures the change in food prices faced by households over time. The food group is the only commodity group of the consumers price index (CPI) for which an index is prepared each month. The CPI is prepared quarterly.
The FPI consists of five subgroups:
• fruit and vegetables
• meat, poultry, and fish
• grocery food
• non-alcoholic beverages
• restaurant meals and ready-to-eat food.
These subgroups are broken down into 14 classes and 17 sections within selected classes.
Since July 2020, 162 different food goods and services are included in the FPI basket (see ‘Definitions’ below). The basket of goods and services is reviewed every three years, with the latest review being implemented in July 2020.
Each item in the basket is allocated an expenditure weight (see ‘Definitions’ below), which shows its importance in a household’s shopping basket relative to other food items. The more that households spend on a particular item the higher the weight assigned to that item is. Items with higher weights have a larger impact on the FPI. For instance, if households spend more on milk than on bread, a 5 percent increase in the price of milk would have a greater impact on the FPI than a 5 percent increase in the price of bread.
It is important to note that the FPI is not a measure of price levels or average prices, it is primarily a measure of price change. A price index is a series of numbers that show how a whole set of prices has changed over time. One index number by itself means nothing. Another index number is needed to compare it with in order to calculate the movement between the two time periods.
en-NZThe FPI makes up 18.72 percent of the CPI and is therefore an important element of quarterly CPI movements.
In terms of limitations, the FPI is a measure of food price change for households only, it should not be used or interpreted as an inflation measure for the economy as a whole.
Economic analysts, the media, researchers, and the National Accounts and Customer Services areas of Statistics NZ.
2006 FPI Review
The 2006 review involved the reselection of the basket of representative food goods and services, reweighting of the basket to better reflect the relative importance of household spending on food based on the 2006/07 HES, and rebasing the index series to June 2006 month (=1000). In addition, the sample of retail outlets that prices are collected from was updated.
A new expenditure classification system, based on the United Nations' Classification of Individual Consumption According to Purpose and adapted to suit New Zealand conditions, was developed by Statistics New Zealand. The new classification system, the New Zealand Household Expenditure Classification (NZHEC), was used for the reweighted FPI and CPI.
Note that care should be taken when comparing the 2002 weights with the 2006 weights. The 2002 weights for the old classification system have been reclassified to the new NZHEC classification. There are some differences in the way the weights were aggregated in 2002 from the way they were aggregated in 2006, due to the use of different classification systems.
Comparing the old and new classification systems at the subgroup level, the NZHEC meat, poultry, and fish subgroup is equivalent to the old system's meat, fish, and poultry subgroup. The NZHEC fruit and vegetables subgroup covers the same items as the old system's subgroup, with the exception of pasta sauces, which are classified under the NZHEC grocery food subgroup. A new subgroup, non-alcoholic beverages, draws on items previously classified under the old system's grocery food subgroup, and the restaurant meals and ready-to-eat food subgroup. Note, takeaway coffee, tea, and milkshakes are classified under the NZHEC restaurant meals and ready-to-eat food subgroup.
The pricing specifications for all the food goods and services in the FPI basket was also reviewed, to ensure that surveyed varieties and sizes are representative of household purchases.
The July 2006 FPI was the first index published on a new index reference of the June 2006 month (=1000). Previous series were published on a base of the June 1999 month (=1000). For categories under the new expenditure classification that had equivalent indexes under the old classification system, the previously published percentage changes for periods up until the June 2006 month were preserved by scaling the index numbers so that the June 2006 month is set to 1000. For categories with no equivalent existing series, a new time series has been calculated back to the June 1999 month.
Price movements for the updated and reweighted basket from the June to July 2006 months were linked to the rescaled FPI at the June 2006 month.
The main source of information used to reweight the June 2006 FPI basket was the 2003/04 HES, which collected detailed information on the spending patterns of nearly 3,000 households. For food items where the HES was not considered to provide accurate information, such as confectionery and soft drinks, alternative information, such as supermarket scan data obtained from the Nielsen Company, was used. The initial weights, for the year to June 2004 (the weight reference period), were 'price updated' to the June 2006 month (the price reference period). This involved expressing the underlying quantities of the weight reference period in the prices of the price reference period.
The FPI was no longer seasonally adjusted from July 2006. Until the 2006 review, fresh fruit and fresh vegetable prices were adjusted to remove the effect of normal seasonal change. From July 2006, seasonally unadjusted prices of fresh fruit and fresh vegetables in the basket were used to calculate the FPI. The ongoing, fully unadjusted FPI was linked to the partly adjusted, previously published FPI at the June 2006 month. This is in line with a recommendation made by the 2004 CPI Revision Advisory Committee to maintain the published time series, even though this may cause some short-term disruption to users in interpreting annual movements.
In addition, the 'elementary aggregate' formula used to calculate lower-level indexes was reviewed. Prior to the 2006 review, the FPI used a 'ratio of arithmetic mean prices' (or Dutot) formula to combine surveyed prices at the first (or elementary) stage of aggregation. For the rebased FPI, the 'geometric mean of price relatives' (or Jevons) formula was introduced for all goods and services in the basket, except for fresh fruit and fresh vegetables.
2008 FPI Review
The 2008 review encompassed the reselection of the basket of representative food goods and services and the reweighting of the basket to reflect the relative importance of household spending on food. A greater level of detail was made available, with both the expenditure weights and index numbers being published for categories at the section level within selected classes.
Food prices rose 8.2 percent in the year to June 2008. This annual increase was the largest since June 1990, when food prices rose 10.0 percent from a year earlier, partly as a result of an increase in the GST rate from 10 percent to 12.5 percent in July 1989.
The pricing specifications of all the food goods and services in the FPI basket were also reviewed to ensure surveyed varieties and sizes are representative of household purchases.
The main source of information used to reweight the FPI basket was the 2006/07 HES, which collected detailed information on the spending patterns of about 2,600 households. However, because the HES doesn't provide accurate information for some food items, such as confectionery and soft drinks, information was also sourced from food manufacturers and distributors, and from supermarket scan data (from the Nielsen Company).
The initial weights for the year to June 2007 (the weight reference period) were 'price updated' to the June 2008 month (the price reference period). This updating involved expressing the underlying quantities of the weight reference period in the prices of the price reference period.
As part of the review of the FPI and the CPI, the regional population weights were updated. These population weights allocate the national expenditure weights of goods and services derived from the HES and other sources to the FPI pricing centers. The population weights ensure that a given price change in Auckland, for example, which then had a population weight of 32.98 percent, would have nearly three times the effect on the national FPI than would the same movement in Christchurch, which had a population weight of 11.55 percent.
The new population weights were calculated using local government boundaries. The 2008 weights were derived by assigning the usually resident population of each regional council area as at June 2007 to the pricing centre(s) within the region. The population weights used previously were based on the usually resident population as at June 2005.
Before the 2008 review, Statistics NZ published the FPI at the subgroup and class levels of the New Zealand Household Expenditure Classification, and for sections within the meat and poultry class. Interest in the FPI had heightened as a result of recent increases in food prices, particularly for dairy and cereal products which lead to expenditure weights and index numbers being published for sections within the milk, cheese, and eggs class and for sections within the bread and cereals class.
2011 FPI Review
The FPI was reviewed in 2011, as part of a three-yearly review of the CPI. The review encompassed the reselection of the basket of representative food goods and services and the reweighting of the basket to reflect the relative importance of household spending on food.
We also reviewed the specifications of all the food products surveyed in the FPI to ensure that surveyed varieties and sizes are representative of household purchases. For packaged foods (such as pasta sauces, jam, and canned fruit), supermarket scan data provided by the Nielsen Company was used to ensure brands being tracked in the FPI align with market shares.
Item pricing specifications were also updated and in some cases the sample of product sizes and varieties reselected. Price collectors were given brand-share targets for selected goods, based on supermarket scan data obtained from the Nielsen Company. This helped to ensure that the mix of brands in the FPI price samples reflects market shares.
The updated FPI sample of products was selected in April 2011. Price collection for the existing and new samples ran alongside each other until June 2011, when collection for the old index ceased.
In addition, we reviewed the selection of supermarkets we collect prices from. Due to an ongoing reduction in the number of distinct supermarket chains, we decided to reduce the number of supermarkets that we collect prices from. Overall, the number of supermarkets tracked decreased by seven to about 70. Before the June 2011 review, about 205,000 food prices were being collected at supermarkets annually. With fewer supermarkets, this decreased to about 190,000 prices.
Food prices were surveyed in 15 urban areas: Whangarei, Auckland, Hamilton, Tauranga, Rotorua, Napier-Hastings, New Plymouth, Wanganui, Palmerston North, Wellington, Nelson, Christchurch, Timaru, Dunedin, and Invercargill.
We used regional population figures to determine the relative importance of each of the 15 areas (each area is allocated a weight, referred to as a regional population weight). This makes sure that price changes at a regional level are accurately reflected in the national FPI. For example, a price change in Auckland (which had 33.43 percent of the population) would have about three times the effect on the national FPI as the same price change in Wellington (which had 11.07 percent of the population).
The 2011 FPI regional population weights were based on the estimated usually resident population at June 2010. The 2008 FPI regional population weights were based on June 2007 population estimates.
Following the Christchurch earthquake on 22 February 2011 price movements for the rest of New Zealand were used to calculate price movements in Christchurch for the March FPI. In June 2011, about half the prices used to calculate the June 2011 FPI had been collected before the 13 June earthquakes, collection was completed on 20 and 21 June, two working days later than other regions where we collect prices for the FPI.
Food price index review: 2014
The 2014 review reselected the food price index basket, and updating the relative importance of the items within it.
We made some minor changes to the specifications of some items as a result of reviewing the basket. However, we did not explicitly review the specifications of food items as part of the 2014 FPI review. Instead, the specifications (eg pack sizes, varieties) of different food items were reviewed as part of the rolling field outlet review, which happened between the 2014 and 2017 reviews.
The relative importance of supermarkets was updated to reflect changes in market shares at the regional level. The importance of supermarkets relative to other store types was also updated for items where prices are also collected in other store types.
The number of regional pricing centres for the FPI was reduced from 15 to 12; Whangarei, Auckland, Hamilton, Tauranga, Napier-Hastings, New Plymouth, Palmerston North, Wellington, Nelson, Christchurch, Dunedin, and Invercargill. Before 1 July 2014, we also collected FPI prices in Timaru, Rotorua, and Wanganui. This was recommended by the CPI Advisory Committee 2013 (recommendation 7) as a means of funding other CPI-related initiatives such as developing the household living-costs price indexes and a seasonally adjusted analytical CPI series.
The 2014 FPI Review was the first to use regional expenditure weights for the five broad regions (Auckland, Wellington, rest of North Island, Canterbury, and rest of South Island), rather than population shares. This change was recommended by the 2013 CPI Advisory Committee (recommendation 6) and aligns with international best practice.
For broad regions with multiple pricing centres (rest of North Island and rest of South Island), we use population shares to allocate the regional expenditure weight to the pricing centres. Previously, we used national expenditure weights in each of the (then) 15 regional pricing centres, weighted by the centre’s population share.
Regional expenditure weights ensure that price changes at a regional level are accurately reflected in the national FPI. For example, a price change in Auckland (which had 33.37 percent of the population and an FPI regional expenditure weight of 35.52 percent) would have had about three times the effect on the national FPI as the same price change in Wellington (which had 11.11 percent of the population and an FPI regional expenditure weight of 11.69 percent).
We calculated regional expenditure weights as proportions of national expenditure (eg 35.52 percent of food expenditure was in the Auckland region) for each FPI class or section (the lowest published level) using HES regional expenditure. We applied class/section level proportions to the individual items within that class or section (eg the regional proportions for fruit were applied to national expenditure on each fruit item) to derive regional expenditure on each individual item (eg spending on apples in Auckland).
Regional expenditure was then expressed in June 2014 prices for the respective region (eg apple expenditure in Auckland was expressed in June 2014 apple prices collected in Auckland). The group-level regional weights were then calculated by aggregating all food expenditure in each broad region.
Food price index review: 2017
Food price index review: 2017 (revised) outlines the changes we made as a result of a review of the food price index (FPI), and implemented in Food price index: October 2017. Reviews of the FPI are generally implemented in July months, however, the 2020 review was delayed a quarter due to the impact of the 2017 Kaikoura earthquake on the work programme. The 2017 review involved reselecting the basket, and updating the relative importance of the items within it.
The 2017 review also included re-setting the FPI indexes to a base period of June 2017 = 1000. Previous series were published on a base of the June 2006 month = 1000. The percentage movements released before the new base period have not changed. This has been achieved by re-scaling the previously published movements and using decimal places to achieve the right percentage changes.
Food price index review: 2020
In August 2020 we introduced the results of the three-yearly review of the food price index with the release of the Food price index: July 2020 results. More information about the review can be found in Food price index review: 2020. The 2020 review involved re-selecting the basket and updating the weights for items in the basket. No items were added or removed from the FPI basket as a result of this review.
Updated weights are based on the 2018/19 Household Economic Survey. These weights will remain in place until the September quarter 2023 CPI release, when we would expect to update them again, using data from the 2021/22 Household Economic Survey: Expenditure. Ordinarily, a three-yearly re-weight of the CPI is sufficient to pick up changing consumer expenditure patterns. With COVID-19, supply and demand factors are likely to speed up this rate of change – with some items more affected than others.
We considered what, if anything, we should do about this; taking on board guidance from international bodies like Eurostat, the International Labour Organization (ILO), and the International Monetary Fund (IMF), talking to other national statistical agencies, and talking to key customers and stakeholders. With industry sources advising that food expenditure patterns had largely returned to normal following the lock-down, we decided to take a conservative approach, and no adjustments were made to weights for any food group items.
However, the lock-down, and inability to send Pricing Officers into stores, did fast forward our plans to use more scanner data in the FPI and CPI, and in the June 2020 month we incorporated point of sale (checkout scanner) price data supplied directly by some supermarkets as an alternative data source. This data replaced all the items that would have previously been manually priced in-store from these supermarkets. This represents a great improvement over in-store price collection, and we intend to continue this method in future.
- Monthly
Papers
|
Price Indexes en-NZ |
Papers
|
|
Food price index review: 2017 (revised) en-NZ |
Papers
|
|
Food price index review: 2020 en-NZ |
Papers
|
|
Consumers price index review: 2020 en-NZ |
Studies
Coverage
Data Collection
Methodology
Food price index review for July 2014
We have reviewed the FPI, as part of a wider, three-yearly consumer price index (CPI) review to ensure the index remains relevant. We updated the basket of representative food items being tracked in the FPI, and updated the relative importance.
The updated weights show that about $37 of every $100 that households spend on food, is spent on grocery food. About $23 is spent on restaurant meals and ready-to-eat food, and about $16 is spent on meat, poultry, and fish. Fruit and vegetable spending accounts for $14, and the remaining $10 is spent on non-alcoholic beverages.
Three items have been added to the FPI basket and none have been removed, taking the number of items in the basket to 166. The items added to the basket are:
- frozen prawns
- packaged leaf salad
- breakfast food drinks.
There were changes to product specifications for chicken pieces and energy drinks. We are now tracking prices for chicken breasts separately from other types of chicken pieces. The weighted average price for 1kg of chicken breast is now included in table 3 of the release. We are also now collecting prices for 250ml containers of energy drinks in supermarkets in addition to 350ml containers in convenience stores. Previously, prices for energy drinks were collected for 350ml in both store types.
Sample size
About 19,000 prices were collected from 560 retail outlets.
Imputation
Due to being unavailable at the time of price collection, on average 0.7 percent of prices (not including seasonal fresh fruit and vegetables) are imputed each month – by carrying forward the previous month’s price.
Reference population
The reference population of the FPI covers approximately 98 percent of the usually-resident New Zealand population living in permanent dwellings. There are no exclusions based on income source or geographic location.
Expenditure weights
Expenditure weights give the relative importance of the food goods and services in the FPI basket.
Expenditure weights are updated every three years as part of regular FPI reviews. The weights are derived largely from the 2012/13 Household Economic Survey (HES). We also used information from food manufacturers and distributors, and supermarket scan data from The Nielsen Company.
FPI weights are based on household spending for the year to June 2013 (the ‘weight reference period’) expressed in June 2014 prices (the ‘price reference period’).
The relative importance of the FPI subgroups shows that about $37 of every $100 that households spend on food is spent on grocery food. About $23 is spent on eating out or takeaways, and about $16 is spent on meat, poultry, and fish. Fruit and vegetables account for $14, and the remaining $10 is spent on non-alcoholic beverages, such as packaged coffee, soft drinks, and juices.
More information on the relative importance of FPI subgroups, classes, and selected sections is in table 6 of this release.
Collection methods
Prices are surveyed by visiting retail outlets in 12 urban areas: Whangarei, Auckland, Hamilton, Tauranga, Napier-Hastings, New Plymouth, Palmerston North, Wellington, Nelson, Christchurch, Dunedin, and Invercargill.
Before 1 July 2014, we also collected FPI prices in Timaru, Rotorua, and Wanganui. However, in line with recommendation 7 of the CPI Advisory Committee 2013, we stopped collecting prices in these three regions, so we could divert the cost of collection towards funding CPI-related initiatives such as household living-costs price indexes and seasonally adjusted analytical CPI series. Price change for these regions will be directly represented by Tauranga, Palmerston North, and Christchurch, respectively.
Fresh fruit and vegetable prices are surveyed weekly, and the remaining food prices are generally surveyed between the 8th and 16th day of the month, although sometimes surveying starts and finishes earlier or later.
Sample design
Food prices are collected from about 560 outlets in the 12 surveyed urban areas. Of these, about 60 are supermarkets, 30 greengrocers, 20 fish shops, 30 butchers, 60 convenience stores (with about half being service stations and the rest being dairies, grocery stores, and superettes), 110 restaurants (for evening meals), and about 250 other suitable outlets (for breakfast, lunch, and takeaway food).
Statistics NZ collects prices from a sample of supermarkets in each of the 12 FPI pricing regions. This sample is designed to be representative of household purchases in each region. It was last reviewed in 2011. The sample of other stores was last reviewed in 2013 as part of the rolling review of outlets.
See CPI rolling review of retail outlets for more information.
Accuracy of the data
Review of the food price index
Reviews of the FPI are undertaken every three years, as part of wider reviews of the consumers price index (CPI). The latest review was implemented with the publication of Food Price Index: July 2014 onwards. The review involved reselecting the basket of representative food goods and services, calculating new national expenditure weights, and moving to regional expenditure weights.
The previous product sample’s final price collection period was June 2014. The updated FPI sample of products also had prices collected in June 2014. An overlapping price collection is necessary when changing a price index, to ensure changes in basket composition (eg basket additions, different outlets) are not reflected as price changes.
See Food price index review: 2014 for more information.
Population weights
From the July 2014 FPI onwards, regional price change is weighted using regional expenditure weights for the five broad regions (Auckland, Wellington, rest of North Island, Canterbury, and rest of South Island). Regional expenditure weights use expenditure in each region to weight regional price change. This ensures that price change in regions where households spend more per person on a particular item relative to other regions (eg Auckland which has 33.37 percent of the population and an FPI regional expenditure weight of 35.52 percent) has more influence on the combined national price change for that item.
For broad regions with multiple pricing centres (rest of North Island and rest of South Island), we use population shares to allocate the regional expenditure weight to the pricing centres.
Previously, we used national expenditure weights in each of the (then) 15 regional pricing centres, weighted by the centre’s population share. This change was recommended by the 2013 CPI Advisory Committee (recommendation 6) and aligns with international best practice.
We calculated regional expenditure weights as proportions of national expenditure (eg 35.52 percent of food expenditure is in the Auckland region) for each FPI class or section (the lowest published level) using HES regional expenditure. We applied class/section level proportions to the individual items within that class or section (eg the regional proportions for fruit was applied to national expenditure on each fruit item) to derive regional expenditure on each individual item (eg spending on apples in Auckland).
Regional expenditure was then expressed in June 2014 prices for the respective region (eg apple expenditure in Auckland was expressed in June 2014 apple prices collected in Auckland). The group level regional weights were then calculated by aggregating all food expenditure in each broad region.
Statistics NZ publishes food and consumers price indexes for five broad regions based on regional council area boundaries. These indexes are available from Infoshare. These regions are Auckland, Wellington, rest of North Island, Canterbury, and rest of South Island.
For the 2014 regional expenditure weights for the five broad regions and 12 regional pricing centres, see table 7 of this release.
Outlet weights
Outlets are given appropriate weights to reflect their relative importance in terms of household spending.
Elementary aggregate formulae
Regional elementary aggregates are calculated for each of the 12 pricing centres from all prices collected for an item within that region. Regional elementary aggregates are calculated using a 'geometric mean of price relatives', or Jevons formula.
The Jevons formula is used to calculate average prices for all food goods and services in the basket, except fresh fruit and fresh vegetables. The Jevons formula assumes that households spend the same amount at each surveyed outlet in each period. This implies that increased quantities are purchased from outlets showing lower-than-average relative price change and decreased quantities from outlets showing higher-than-average price change. The calculation of fresh fruit and vegetable average prices uses the Dutot formula.
Information about the Food Price Index gives more information on the Jevons and Dutot formulae (see elementary aggregate formulae).
'On special' prices
Items that are 'on special' are included in the FPI at the price levels observed at the time of price collection. Quantity specials (for example, three loaves of bread for $5.00) are also taken into account (as the price per loaf for the special is usually lower than the price of a single loaf). Where discounted prices are available only to customers who belong to discount schemes, this is represented in the FPI by collecting these prices at some outlets within a region, but not others.
Consistency with other periods or datasets
Impact of the Christchurch earthquakes on price collection
Following the Christchurch earthquake on 22 February 2011 price movements for the rest of New Zealand were used to calculate price movements in Christchurch for the March 2011 FPI. About half the prices used to calculate the June 2011 FPI had been collected before the 13 June earthquakes; collection was completed on 20 and 21 June, two working days later than other regions where we collect prices for the FPI.
Index base
The FPI has an index reference period of the June 2006 month (=1000). This is the benchmark to which prices in other periods are compared (eg if the index number in a later period is 1150, prices have increased by 15.0 percent since the index reference period). Prices for later periods can also be compared in the same fashion.
Seasonal adjustment of prices – fresh fruit and vegetables
Until the June 2006 month, fresh fruit and vegetable items that exhibited a seasonal pattern were adjusted to remove the effect of normal seasonal change. From the July 2006 month onwards, the FPI incorporates seasonally unadjusted prices for fresh fruit and vegetables. This change is in line with a recommendation made by the 2004 CPI Revision Advisory Committee.
The ongoing, fully unadjusted FPI is linked at the June 2006 month to the previously published FPI, which is partly seasonally adjusted. As such, care is required when comparing annual movements over this transition period. Annual movements calculated over the annual period encompassing the June 2006 month were based on fully unadjusted index numbers for the latest month, compared with adjusted index numbers for fresh fruit and vegetables for the same month of the previous year.
Reconciling the FPI and food group of the CPI
When comparing the FPI and the food group of the CPI, strictly speaking, the quarterly food group index number is not the average of the relevant three monthly FPI numbers. There are some technical differences between the monthly FPI indexes and quarterly indexes.
See Food prices in the consumers price index and food price index for more information.
Interpreting the data
Seasonal adjustment The 2013 CPI Advisory Committee recommended we add analytical seasonally adjusted series to our publications. We are seasonally adjusting the CPI and FPI at the all groups, group, subgroup, and class levels. The headline FPI will remain unadjusted. We have seasonally adjusted using direct adjustment rather than indirect since this produced better quality statistics. Indirect seasonal adjustment occurs when individual component series of the main aggregate series are seasonally adjusted, then aggregated to derive totals. For example, an indirect seasonally adjusted fruit series would be compiled by adding all the seasonally adjusted series (for apples, pears, kiwifruit, etc) together. Direct seasonal adjustment occurs when seasonally adjustment is done at the aggregate level, independently of seasonally adjusting the components. A direct seasonally adjusted fruit series would be made up by adjusting the aggregate of all the unadjusted series (for apples, pears, kiwifruit, etc). We use the x13 ARIMA-SEATS package to run our seasonal adjustment. For more information about seasonal adjustment see Seasonal adjustment in Statistics New Zealand and for how it relates to the CPI see Price Index News: CPI sources and methods articles.
See Analytical comsumer price index seasonally adjusted series for further information.
Seasonal availability of fruit and vegetables
Fruit and vegetable prices are reflected in the FPI when there is enough produce available to estimate representative average prices. For example, prices for nectarines are historically not included in the April and May FPI. Similarly, prices for strawberries are not included in the May and June FPI. This is because not enough prices can be collected from stores during these months. No price change is shown in the FPI for these items during these months. When produce returns to sufficient levels, the prices are again reflected in the FPI. Price movements then reflect the price change from the month that the item was last included to the current month.
Weighted average retail prices of selected food items
Table 3 contains a selection of weighted average retail prices for the current and previous months. These weighted average retail prices were calculated from prices collected in the June 2006 month. Subsequent months' weighted average prices are then calculated by applying price index movements for the relevant items. These are not statistically accurate measures of average transaction price levels, but are reliable indicators of percentage changes in prices.
en-NZCoverage
New Zealand
en-NZFood Price Index Series 2017
Methodology
Food price index review
We have reviewed the FPI, as part of a wider, three-yearly consumer price index (CPI) review to ensure the index remains relevant. The latest review was implemented with the publication of Food Price Index: October 2017 month onwards. FPI reviews are generally implemented in July months, however implementation the 2017 review was delayed a quarter due to the impact of the 2016 Kaikoura earthquakes on the work programme.
We updated the basket of representative food items being tracked in the FPI, updated the relative importance (weights) of the items in the basket and re-set the FPI indexes to a base period of June 2017 = 1000. Previous series were published on a base of the June 2006 month = 1000.
The updated weights show that about $34 of every $100 that households spend on food, is spent on grocery food. About $26 is spent on restaurant meals and ready-to-eat food, and about $15 is spent on meat, poultry, and fish. Fruit and vegetable spending accounts for $15, and the remaining $10 is spent on non-alcoholic beverages.
Three items have been added to the FPI basket and seven have been removed, taking the number of items in the basket to 162. The items added to the basket are:
• fresh herbs
• olives
• flavoured tea bags.
The removed basket items are: - alfalfa sprouts - spring onions - taro - Canned corn - luncheon sausage - cottage cheese - takeaway milkshakes.
See Food price index review: 2017 (revised) for more information.
Sample size
About 19,000 prices were collected from 560 retail outlets.
Imputation
Due to being unavailable at the time of price collection, on average 0.7 percent of prices (not including seasonal fresh fruit and vegetables) are imputed each month – by carrying forward the previous month’s price.
Reference population
The reference population of the FPI covers approximately 98 percent of the usually-resident New Zealand population living in permanent dwellings. There are no exclusions based on income source or geographic location.
Expenditure weights
Expenditure weights give the relative importance of the food goods and services in the FPI basket.
Expenditure weights are updated every three years as part of regular FPI reviews. The weights are derived largely from the 2012/13 Household Economic Survey (HES). We also used information from food manufacturers and distributors, and supermarket scan data from The Nielsen Company.
FPI weights are based on household spending for the year to June 2016 (the ‘weight reference period’) expressed in September 2017 prices (the ‘price reference period’).
More information on the relative importance of FPI subgroups, classes, and selected sections can be found in table 6 of each food price releases’ tables.
Collection methods
Prices are surveyed by visiting retail outlets in 12 urban areas: Whangarei, Auckland, Hamilton, Tauranga, Napier-Hastings, New Plymouth, Palmerston North, Wellington, Nelson, Christchurch, Dunedin, and Invercargill.
Fresh fruit and vegetable prices are surveyed weekly, and the remaining food prices are generally surveyed between the 8th and 16th day of the month, although sometimes surveying starts and finishes earlier or later.
Sample design
Food prices are collected from about 560 outlets in the 12 surveyed urban areas. Of these, about 60 are supermarkets, 30 greengrocers, 20 fish shops, 30 butchers, 60 convenience stores (with about half being service stations and the rest being dairies, grocery stores, and superettes), 110 restaurants (for evening meals), and about 250 other suitable outlets (for breakfast, lunch, and takeaway food).
Statistics NZ collects prices from a sample of supermarkets in each of the 12 FPI pricing regions. This sample is designed to be representative of household purchases in each region. It was last reviewed in 2011. The sample of other stores was last reviewed in 2013 as part of the rolling review of outlets.
See CPI rolling review of retail outlets for more information.
Accuracy of the data
Population weights
From the July 2014 FPI onwards, regional price change is weighted using regional expenditure weights for the five broad regions (Auckland, Wellington, rest of North Island, Canterbury, and rest of South Island). This ensures that price change in regions where households spend more per person on a particular item relative to other regions. For broad regions with multiple pricing centres (rest of North Island and rest of South Island), we use population shares to allocate the regional expenditure weight to the pricing centres.
Previously, we used national expenditure weights in each of the (then) 15 regional pricing centres, weighted by the centre’s population share. This change was recommended by the 2013 CPI Advisory Committee (recommendation 6) and aligns with international best practice.
Outlet weights
Outlets are given appropriate weights to reflect their relative importance in terms of household spending.
Elementary aggregate formulae
Regional elementary aggregates are calculated for each of the 12 pricing centres from all prices collected for an item within that region. Regional elementary aggregates are calculated using a 'geometric mean of price relatives', or Jevons formula.
The Jevons formula is used to calculate average prices for all food goods and services in the basket, except fresh fruit and fresh vegetables. The Jevons formula assumes that households spend the same amount at each surveyed outlet in each period. This implies that increased quantities are purchased from outlets showing lower-than-average relative price change and decreased quantities from outlets showing higher-than-average price change. The calculation of fresh fruit and vegetable average prices uses the Dutot formula.
Information about the Food Price Index gives more information on the Jevons and Dutot formulae (see elementary aggregate formulae).
'On special' prices
Items that are 'on special' are included in the FPI at the price levels observed at the time of price collection. Quantity specials (for example, three loaves of bread for $5.00) are also taken into account (as the price per loaf for the special is usually lower than the price of a single loaf). Where discounted prices are available only to customers who belong to discount schemes, this is represented in the FPI by collecting these prices at some outlets within a region, but not others.
Consistency with other periods or datasets
Index base
The FPI now has an index reference period of the June 2017 month (=1000). This is the benchmark to which prices in other periods are compared (eg if the index number in a later period is 1150, prices have increased by 15.0 percent since the index reference period). Prices for later periods can also be compared in the same fashion. The previous reference period was June 2006 month (=1000). This was done so that series that had fallen to 2 digits would have a higher level of clarity. For example, an index that moves from 11 to 10 requires a 9 percent fall in price, compared to an index moving from 1000 to 999 that only needs a 0.1 percent change. Due to the rebase date and implementation date being different, a few periods after the rebase have decimal places. This is to ensure the percent changes released remain consistent.
Seasonal adjustment of prices – fresh fruit and vegetables
Until the June 2006 month, fresh fruit and vegetable items that exhibited a seasonal pattern were adjusted to remove the effect of normal seasonal change. From the July 2006 month onwards, the FPI incorporates seasonally unadjusted prices for fresh fruit and vegetables.
This change is in line with a recommendation made by the 2004 CPI Revision Advisory Committee.
The ongoing, fully unadjusted FPI is linked at the June 2006 month to the previously published FPI, which is partly seasonally adjusted. As such, care is required when comparing annual movements over this transition period. Annual movements calculated over the annual period encompassing the June 2006 month were based on fully unadjusted index numbers for the latest month, compared with adjusted index numbers for fresh fruit and vegetables for the same month of the previous year.
Reconciling the FPI and food group of the CPI
When comparing the FPI and the food group of the CPI over a review period, note that the quarterly food group index number is not the average of the relevant three-monthly FPI numbers. Where there are changes to food basket items, prices for new CPI basket items would be collected in April, May and June (to apply the price change between the June and September quarters), whereas prices for new FPI items are collected for June (to apply the price change between June and July months).
The FPI has a monthly price reference period, and the prices for the June month can differ from the average of the April, May and June months. As a result, the monthly and quarterly base weights can differ, even though the same annual quantities were used.
Interpreting the data
Seasonal adjustment
The 2013 CPI Advisory Committee recommended we add analytical seasonally adjusted series to our publications. We produce an analytical series where the FPI (and CPI) are seasonally adjusted at the all groups, group, subgroup, and class levels. The headline FPI remains unadjusted. We have seasonally adjusted using direct adjustment rather than indirect since this produced better quality statistics. Indirect seasonal adjustment occurs when individual component series of the main aggregate series are seasonally adjusted, then aggregated to derive totals. For example, an indirect seasonally adjusted fruit series would be compiled by adding all the seasonally adjusted series (for apples, pears, kiwifruit, etc) together. Direct seasonal adjustment occurs when seasonally adjustment is done at the aggregate level, independently of seasonally adjusting the components. A direct seasonally adjusted fruit series would be made up by adjusting the aggregate of all the unadjusted series (for apples, pears, kiwifruit, etc). We use the x13 ARIMA-SEATS package to run our seasonal adjustment.
Seasonal availability of fruit and vegetables
Fruit and vegetable prices are reflected in the FPI when there is enough produce available to estimate representative average prices. For example, prices for nectarines are historically not included in the April and May FPI. Similarly, prices for strawberries are not included in the May and June FPI. This is because not enough prices can be collected from stores during these months. No price change is shown in the FPI for these items during these months. When produce returns to sufficient levels, the prices are again reflected in the FPI. Price movements then reflect the price change from the month that the item was last included to the current month.
Weighted average retail prices of selected food items
Table 3 contains a selection of weighted average retail prices for the current and previous months. These weighted average retail prices were calculated from prices collected in the June 2006 month. Subsequent months' weighted average prices are then calculated by applying price index movements for the relevant items. These are not statistically accurate measures of average transaction price levels, but are reliable indicators of percentage changes in prices.
en-NZFood Price Index Series 2020
Methodology
Review of the FPI
Reviews of the FPI are generally undertaken every three years, as part of a wider review of the consumers price index (CPI). We implemented the latest review when the July 2020 FPI was published. The review involved reselecting the basket of representative goods and services, updating the new national expenditure weights, and updating regional expenditure weights. No items were added or removed from the FPI basket as a consequence of the latest review. More information on the updated expenditure weights can be found below or in Food price index review: 2020.
Sample size
About 19,000 prices are collected from 560 retail outlets.
Imputation
Due to being unavailable at the time of price collection, on average less than 1 percent of prices (not including seasonal fresh fruit and vegetables) are imputed each month – using class mean imputation.
Reference population
The reference population of the FPI covers approximately 98 percent of the usually-resident New Zealand population living in permanent dwellings.
Expenditure weights
Expenditure weights give the relative importance of the food goods and services in the FPI basket.
As noted above, expenditure weights are updated every three years as part of regular FPI reviews. The current set of weights are derived from the 2018/19 Household Economic Survey (HES) and other sources. We also used information from food manufacturers and distributors, and supermarket scan data from The Nielsen Company.
FPI weights are based on household spending for the year to June 2020 (the ‘weight reference period’), expressed in June 2020 quarter prices (the ‘price reference period’).
The updated weights show that about $34 of every $100 that households spend on food, is spent on grocery food. About $27 is spent on restaurant meals and ready-to-eat food, and about $15 is spent on meat, poultry, and fish. Fruit and vegetable spending accounts for $13, and another $10 is spent on non-alcoholic beverages. (Percentages do not add to 100 due to rounding)
The CPI Review tables [Food price index review: 2020]https://www.stats.govt.nz/assets/Uploads/Food-price-index/Food-price-index-July-2020/Download-data/food-price-index-rewiew-2020.xlsx)show the relative importance of CPI groups, subgroups, and classes.
Impacts of COVID-19 on the 2020 CPI and FPI re-weights
As noted in Impacts of COVID-19 on the 2020 CPI re-weight we did not make any adjustments to weights in the food group as a part of the three-yearly review of the FPI due to the impact of COVID-19. However, our intention to adjust the intend adjust the weights for domestic and international airfares and overseas accommodation cost pre-paid in New Zealand means that the relative weight of all other CPI basket items, including food, will scale in association with the annual reweight of these three items. We aim to keep an eye on shifting expenditure patterns for other expenditure items over the next three years’, but would only consider changing weights for other items where there is a clear-cut case for doing so. This would be signalled well in advance of any change.
Collection methods
Prices are surveyed by visiting retail outlets in 12 pricing centres: Whangarei, Auckland, Hamilton, Tauranga, Napier-Hastings, New Plymouth, Palmerston North, Wellington, Nelson, Christchurch, Dunedin, and Invercargill.
From the April 2020 FPI release onwards, we have incorporated point of sale (checkout scanner) price data supplied directly by some supermarkets as an alternative data source. This data replaced all the items that would have previously been manually priced in-store from these supermarkets. Rather than being a point-in-time reference, the data feeding into the FPI is a weighted average for each month. This represents a great improvement over in-store price collection, and we intend to continue this method in future.
For other outlets, fresh fruit and vegetable prices are surveyed weekly, and the remaining food prices are generally surveyed between the 8th and 16th day of the month, although sometimes surveying starts and finishes earlier or later.
Sample design
Food prices are collected from close to 560 outlets in the 12 pricing centres. Of these, 56 are supermarkets, 32 are greengrocers, 53 are fish shops and butchers, 68 are convenience stores (eg dairies and service stations), and 350 are restaurants (with and without bars) and takeaways.
Accuracy of the data
Elementary aggregate formula
Average prices in the FPI are called elementary aggregates. These elementary aggregates are the first level of the index aggregation. We calculate regional elementary aggregates for each of the 12 pricing centres where price collection supports regional estimation. In other cases, we calculate regional elementary aggregates for five CPI broad regions (Auckland, Wellington, rest of North Island, Canterbury, rest of South Island) or, where prices do not support regional estimation, directly to a national elementary aggregate. Since the 2006 review of the CPI, we have used the geometric mean, or Jevons, formula to calculate the elementary aggregate indexes for items where outlet substitution is possible (eg for groceries and appliances).
We use the 'ratio of arithmetic mean prices', or Dutot, formula for fresh fruit and vegetables (as the first stage of aggregation is across both outlets within each region, and across weeks within each month).
Regional expenditure weights
From the September 2014 quarter CPI onwards, we weight regional price change using regional expenditure weights for the five broad regions (Auckland, Wellington, rest of North Island, Canterbury, and rest of South Island). Regional expenditure weights use expenditure in each region to weight regional price change. This ensures that price change in regions where households spend more per person on a particular item relative to regions (eg Auckland has 33.41 percent of the population and a FPI regional expenditure weight of 34.85 percent) has more influence on the combined national price change for that item.
For broad regions with multiple pricing centres (ie rest of North Island and rest of South Island), we use population shares to allocate the regional expenditure weight to the pricing centres.
We publish FPIs for the five broad regions based on regional council area boundaries. These indexes are available from Infoshare.
Regional expenditure weights for the five broad regions can be found in the FPI review tables Food price index review: 2020.
Outlet weights
We give outlets appropriate weights to reflect their relative importance in terms of household spending.
'On special' prices
As noted above, price data for some supermarkets is now a weighted average for the month, which reduces the ability to identify ‘special’ prices. For other outlets, items that are 'on special' are included in the FPI at the price levels observed at the time of price collection. Quantity specials (eg 15-pack of beer at a cheaper shelf price than the 12-pack) are also considered where appropriate (as the price per bottle for the special is lower).
Consistency with other periods or datasets
Index base
The FPI has an index reference period of the June 2017 month (=1000). This is the benchmark to which prices in other periods are compared (eg if the index number in a later period is 1150, prices have increased by 15.0 percent since the index reference period). Prices for later periods can also be compared in the same fashion.
Seasonal adjustment of prices – fresh fruit and vegetables
Until the June 2006 month, fresh fruit and vegetable items that exhibited a seasonal pattern were adjusted to remove the effect of normal seasonal change. From the July 2006 month onwards, the FPI incorporates seasonally unadjusted prices for fresh fruit and vegetables. This change is in line with a recommendation made by the 2004 CPI Revision Advisory Committee.
The ongoing, fully unadjusted FPI is linked at the June 2006 month to the previously published FPI, which is partly seasonally adjusted. As such, care is required when comparing annual movements over this transition period. Annual movements calculated over the annual period encompassing the June 2006 month were based on fully unadjusted index numbers for the latest month, compared with adjusted index numbers for fresh fruit and vegetables for the same month of the previous year.
Rather than seasonally adjusting fruit and vegetables, the prices of these items are reflected in the FPI when there is enough produce available to estimate representative average prices. For example, prices for nectarines are historically not included in the April and May FPI. Similarly, prices for strawberries are not included in the May and June FPI. This is because not enough prices can be collected from stores during these months. No price change is shown in the FPI for these items during these months. When produce returns to sufficient levels, the prices are again reflected in the FPI. Price movements then reflect the price change from the month that the item was last included to the current month.
However, we do produce an analytical seasonally adjusted series for the FPI (and CPI) at the all groups, group, subgroup, and class levels. The headline FPIs remain unadjusted. We have seasonally adjusted using direct adjustment rather than indirect since this produced better quality statistics.
Indirect seasonal adjustment occurs when individual component series of the main aggregate series are seasonally adjusted, then aggregated to derive totals. For example, an indirect seasonally adjusted fruit series would be compiled by adding all the seasonally adjusted series (for apples, pears, kiwifruit, etc) together.
Direct seasonal adjustment occurs when seasonally adjustment is done at the aggregate level, independently of seasonally adjusting the components. A direct seasonally adjusted fruit series would be made up by adjusting the aggregate of all the unadjusted series (for apples, pears, kiwifruit, etc). We use the x13 ARIMA-SEATS package to run our seasonal adjustment.
Reconciling the FPI and food group of the CPI
When comparing the FPI and the food group of the CPI over a review period, note that the quarterly food group index number is not the average of the relevant three-monthly FPI numbers. Where there are changes to food basket items, prices for new CPI basket items would be collected in April, May and June (to apply the price change between the June and September quarters), whereas prices for new FPI items are collected for June (to apply the price change between June and July months).
The FPI has a monthly price reference period, and the prices for the June month can differ from the average of the April, May and June months. As a result, the monthly and quarterly base weights can differ, even though the same annual quantities were used.
Interpreting the data
Weighted average retail prices of selected food items
Table 3 of FPI releases contains a selection of weighted average retail prices for the current and previous months. These weighted average retail prices were calculated by applying index movements to weighted average prices for the June 2017 quarter CPI. They are not statistically accurate measures of average transaction price levels, but do provide a reliable indicator of percentage changes in prices.
Timing of published data
We generally publish the FPI 10 working days after the reference month.
For more information on the review, please contact:
Fiona Smillie or James Griffin 04 931 4600 info@stats.govt.nz
en-NZ