How new car price changes are measured in the CPI

Purchase of new motor cars had an expenditure weight of 1.69 percent of the Consumers Price Index (CPI) as at the June 2006 quarter. One of the challenges faced in measuring new car prices is that the quality of new cars does not remain constant over time. This presents an issue as the CPI should reflect only "pure" price changes.

Position in CPI structure

The following table shows that the new motor cars class falls within the transport group of the New Zealand Household Expenditure Classification used for the CPI.

 

The expenditure weight for new motor cars was derived using independent data sources. Information from the Household Economic Survey (HES), the most common source of CPI weights, was not used because it was considered less reliable given that new cars are relatively infrequently purchased, thus resulting in a high sampling error.

The method used to estimate the current expenditure weight for purchase of new cars is as follows:

• Registration data for new motor cars (disaggregated by quarter and by make and model) for the year to March 2006 was obtained. Under normal circumstances, information for the year to June 2004 would have been used (as this is the weight reference period). However, analysis of the data indicated a significant shift from large cars to smaller cars after the weight reference period. Statistics New Zealand therefore decided to reflect this shift.
• The new-car registration data was filtered to align as closely as possible to the CPI target population of private households living in permanent private dwellings. This means that only cars registered for private use by individuals were included, as illustrated by the diagram below:

 

• The cars were then split into three groups based on engine sizes: small (less than 1750 cc), medium (1750 cc to 2549 cc) and large (2550 cc and above). This helps to ensure that cars in each engine size category are properly represented.
• The registration data was combined with list prices at the model level (which are publicly available) to derive annual expenditure values for the three engine-size categories. Adjustments were then made to exclude out-of-scope expenditure (such as those by international visitors to New Zealand) and to reflect estimated discounts for cash.
• Finally, expenditures for the three engine-size categories were allocated to the sampled makes/models within each engine size category, based on the registration data.

The expenditure weight allocated to the purchase of new cars represents gross expenditure on new cars for private use by households. The expenditure weight for purchase of second-hand cars, derived from the HES, represents net purchases of second-hand cars (that is, purchases of second-hand cars minus sales and trade-ins of second-hand cars, including trade-ins on purchases of new cars). The June 2006 quarter weight for the purchase of second-hand motor cars (3.30 percent) is higher than that for the purchase of new motor cars (1.69 percent).

Price collection

The CPI tracks the price changes of a sample of new car makes and models. This sample was reselected as part of the 2006 CPI review to reflect changes in the vehicles households were buying. For example, the current sample has a greater representation of four-wheel drive, diesel-powered and luxury European vehicles compared with the previous sample.

Prices for new cars are collected by price collectors visiting a sample of authorised car dealers in the middle month of each quarter. The current sample consists of about 115 new car dealers located in the 15 CPI urban areas. Each quarter, respondents are asked to provide the prices of selected models. For practical reasons, only discounts which are generally available to the public and are not subject to individual negotiation are taken into account, including publicly advertised promotions, distributor-backed discounts and cash-back offers. Respondents are also asked if the price provided includes on-road costs.

Statistics New Zealand also collects other information from the national head offices of distributors who supply vehicles to local dealers (from here on referred to as distributors). See sample maintenance and quality adjustment for details of the information collected from head offices.

Sample maintenance

At times, respondents in the sample may need to be replaced (for example, due to closure of business). When this happens, a replacement authorised dealer is selected from the same area. In cases where a particular region does not have an authorised dealer, prices from a neighbouring region are used.

In order to maintain the sample of new cars, changes to the sampled models need to be taken into account. Each quarter, the head offices of car distributors that have samples in the survey are contacted and asked whether there have been any changes to the models in the sample or whether any of the models have been discontinued.

If a sampled model is updated or replaced, Statistics NZ moves to surveying the replacement model in place of the old one immediately, even if the old model is still available. If a model in the sample is discontinued, the car distributor is asked what the most appropriate replacement model would be for a customer who would previously have purchased the old model.

Whenever a model is discontinued or changed, a quality adjustment is made to remove the effect of any changes in features (see quality adjustment).

Quality adjustment

One of the challenges in measuring new car prices is that the quality of new cars does not remain constant over time. For instance, a manufacturer might replace an existing model with a new one that has additional features as well as a more powerful engine. This presents an issue because the CPI should reflect only "pure" price changes not affected by changes in quality. The procedure for handling quality adjustment for new cars is discussed in the following section.

Quality adjustment: perceived value

Each quarter, the head offices of car distributors that have samples in the survey are contacted and asked if there have been any changes to the models in their sample, including the addition or removal of features, or whether any of the models have been discontinued entirely. Both situations can be treated in the same way in terms of the quality assessment that is carried out. The procedure is as follows:

• First, a comparison of the old model and the updated (or replacement) model is made using detailed specifications in brochures obtained from the head office of the appropriate car distributor, and all the new, altered or removed features are identified.
• Once changes have been identified, the distributor is asked to provide the "perceived" dollar value to a typical customer of each new, altered or removed feature. Perceived values for engine changes are not asked of the distributor, as these are directly adjusted by Statistics NZ.
• Once perceived values have been obtained, they are compared with the values used for previous new car assessments to ensure they are broadly consistent with previous assessments. If not, alterations to the values may be made to ensure consistency with previous assessments.
• The perceived values are either added to (for removed features) or subtracted from (for added features or other improvements) the current price of the new model, to remove the effect of the change in quality from the previous period.

Quality adjustment: engine quality

The value of any change in engine quality is estimated using information on changes in maximum power and maximum torque, and the revolutions per minute (RPM) at which they occur. The relative weights assigned to power and to torque, and the ratio used for engine value to car value are based on historic discussions with, and information obtained from, a number of distributors.

The estimated value derived from the above information, which may be positive (in the case of an increase in engine quality) or negative (for engine quality decrease), is then subtracted from the current price of the new model to remove the effect of the change in quality from the previous period.

It should be noted that engine quality changes which are unrelated to power and torque (such as fuel economy and environmental friendliness), are not currently taken into account. This is due to the difficulty in independently assessing and quantifying such changes.

Estimation

Using the information collected and after adjusting for quality changes, regional average prices are calculated for all 15 regions for each surveyed model. Price movements in these regional average prices are then combined using expenditure weights based on regional population shares to derive the national new motor cars index.

Back to Price Index News: October 2007