Workshop on Price Index Compilation Issues

February23-27, 2015

Quality Change and New Goods

Gefinor Rotana Hotel, Beirut, Lebanon

Lecture Outline

The Quality Change ProblemWhy are Quality Adjustments (QA’s) Important?When Should QA’s Be Made?Methods for Estimating the Value of the Quality DifferenceSource of QA Data

Lecture Outline

The New Goods ProblemUse of a Fixed-Base Price Index New Goods Offer Purchasers New Possibilities

Three problems:

A permanently missing item problem - quality adjustmentA sampling problem – new models are introduced and old ones disappearA new products problem

The Quality Change Problem

Why are Quality Adjustments (QA’s) Important?

Purpose of the index• measure of pure price change, and thus,

inflation in the sector of the economy covered by the index

• deflator to remove effects of price changes

• adjustments to economic payments to offset the effects of inflation

Why are Quality Adjustments (QA’s) Important?

Separate price changes from volume changes changes in quality provide more utility and additional satisfaction to usersas such, they represent a change in volumeany increase in value due to quality change should be reflected as a change in volumeif quality change is not removed, it will be reflected as a price change

Why are Quality Adjustments (QA’s) Important?

Affects adjustments made using the price index

if quality is increasing (decreasing), but not removed from the index, it will overstate (understate) price changeany adjustments made using the index (e.g., as deflators) will result in understated volume changes if the quality change is positivealternatively, any adjustments to payments (e.g., cost of living increases) will result in overstated price changes if the quality change is positive

When Should QA’s Be Made?

Sampled products permanently disappear and are replacedAttempt to obtain product with the same characteristics so that we can measure pure price change in our price indexMust make a determination if replacement product has a different level of qualityIf it does, an adjustment in price is needed which reflects the amount of the quality difference

Price statisticians do take account of quality changes

They use the matched models method:

sampled models (varieties) selected using detailed product descriptions on “initiation”prices are recorded in initial month, and monitored in subsequent monthslike is compared with like.

Differentiated items with high rates of model turnover:

jan feb mar apr may jun

A x x x x x xB x x xC x x xD x x x xE x x x x x xF x x x

Alternative methods of quality adjustment

Implicit methods:overall mean/targeted mean imputationcomparable replacementoverlapcarry-forward

Explicit methods: expert judgment quantity adjustmentdifferences in production/option costs hedonic approach

Even if you do nothing you make an implicit quality adjustment.

Overall mean/targeted mean imputation

The overall mean imputation is the computationally straightforward. It is based on the assumption of similar price movements. A targeted form of the method would use similar price movements - sample size.The class mean imputed price changes are based on products whose replacement price has benefited from a quality adjustment or the new replacement product has been judged to be directly comparable. The imputation bias depends on the ratio of unavailable values and the difference between the mean price change for existing products and missing ones.

Comparable replacement

The method relies on the efficacy of the respondents and, in turn, on the adequacy of the specifications used as a description of the price basis. There is an incentive to assume replacements are comparable.

Overlap method

Measures the ratio, in a common overlap period, of the prices of the old and replacement product prices. This is taken to be an indicator of their quality differences. The assumption is that the quality difference in any period equates to the price difference at the time of the splice. The timing of the switch is thus crucial – product life cycles. The quality difference is not related to technical specifications or resource-costs, but to the relative prices. The overlap method is implicitly employed when samples of products are rotated.

Carry forward

Induces undue stability into the index, especially for high inflation countries.

Expert opinion

Industry specialistsDelphi method Objective methods much preferred.

Quantity adjustment

Such scaling is most important. It should not be the case that, for example, because an industrial lubricant is now sold in 5 litre containers instead of 2.5 litre ones, that prices have doubled. Production of a bottle of 100 pills each having 50 milligrams of a drug may not be comparable to a bottle of 50 pills of 100 milligrams, even though both bottles contain 5,000 milligrams of the same drug - non-linearities.

sizepp tm

tm ˆ

Differences in production/option costs

Production costs - particularly appropriate - exclude mark ups and indirect taxesOption costs - the cost of producing something as standard may be lower than when it was an option.

- the option may be valued at say an additional x when sold separately. However, when it is sold as standard many of the purchasers will not value it so highly, since these were the very ones who chose the standard one. – otherwise similar to quantity adjustment:

zpp tm

tm ˆ

Hedonics

i

K

kkiki zp

101n

A set of (zk = 1,….,K) characteristics of the models are identified and data over i=1,…,N models are collected. A hedonic regression of the (log) price of model i, pi , on its set of quality characteristics zki is given by:

Figure 7.2, Scatter diagram

0

500

1000

1500

2000

2500

3000

3500

0 200 400 600 800 1000 1200

Speed, MHz.

Pri

ce, £

s st

erlin

g820.0 Speed261.3436.658riceP̂ 2 R

Dependent variable Price ln Price Constant -725.996 (2.71)** 6.213 (41.95)*** Speed (Processor, MHz) 2.731 (9.98)*** 0.001364 (9.02)*** RAM (random access memory – MB) 1.213 (5.61) *** 0.000598 (5.00) *** HD (hard drive capacity – MB) 4.517 (1.96)* 0.003524 (2.76)** Brand (benchmark: Compaq Deskpro) Compaq Presario -199.506 (1.89)* -0.152 (2.60)** Compaq Prosignia -180.512 (1.38)* -0.167 (2.32)* Dell -1,330.784 (3.74)*** -0.691 (3.52)*** Processor (benchmark: AMD Athlon) Intel Celeron 393.325 (4.38)*** 0.121 (2.43)** Intel Pentium III 282.783 (4.28)*** 0.134 (3.66)*** Rom-drive (benchmark: CD-ROM)† CD-RW (compact disk-rewritable) 122.478 (56.07)*** 0.08916 (2.88)** DVD (digital video drive) 85.539 (1.54) 0.06092 (1.99)* Dell* Speed (MHz) 1.714 (4.038)*** 0.000820 (3.49)*** N 63 63

2R 0.934 0.934

† Read only memory. Figures in brackets are t-statistics testing a null hypothesis of the coefficient being zero. ***,** and * denote statistically significant at a 0.1 percent, 1 percent and 5 percent level respectively, tests being one-tailed.

Sample space and item replacement/substitution

The matching of prices of identical items over time, by its nature, is likely to lead to the monitoring of a sample of items increasingly unrepresentative of the population of transactions. Respondents may keep with their selected items until they are no longer produced, i.e., continue to monitor old items with unusual price changes and limited sales. Yet on item replacement, respondents may select unpopular comparable items to avoid explicit quality adjustments; obsolete items with unusual price changes are replaced by near obsolete items with again, unusual price changes. The substitution of an item with relatively high sales for an obsolete one has its own problems, since the difference in quality is likely to be substantial and substantive, beyond that which can be attributed to, say, the price difference in some overlap period.

The New Goods Problem

Use of a Fixed-Base Price Index A fixed basket does not allow for the immediate inclusion of new goodsNew products are continually introduced and some are introduced through product replacementsSuch replacements are not timely because the new product most likely has been in the market place for awhile

The New Goods Problem

The market life-cycle for new goods is that they are introduced at initially high prices and then proceed to decline in price over timeThese price declines are not captured in the price indices resulting in an upward bias

New Goods Offer Purchasers New Possibilities

New goods often result in the development of new applications that were not previously usedThey also give rise to many complementary products and innovations that make them more efficientThis affects the supply and demand for both the new products and their complementary products

Possible Solutions to the New Goods Problem

More frequent review and updates to the weight structure and product samplesEstimation and aggregation structure that allows for introduction of new productsSample rotation and augmentation

Table 8.1, Sample Augmentation Example Products Base

weight Revised weight

Period 1

Period 2 Period 3 Period 4

A 0.6 0.5 100.00 101.00 101.50 102.50 B 0.4 0.3 100.00 102.00 102.50 103.00 All items 0.8 100.00 101.40 101.90 102.70 C 100.00 98.50 98.00 Spliced C 0.2 100.00 101.40 99.88 99.37 Revised all items

100.00 101.40 101.50 102.05

Information Requirements for a Strategy for Quality Adjustment: statistical metadata.

Statistical agencies should monitor the incidence of missing items and if the incidence is high, then at a more detailed level of the system. Where the incidence is high the ratios of temporary missing prices, comparable replacements and non-comparable replacements to the overall number of prices, and the methods for dealing with each of these three circumstances should also be monitored. Product specific information such as the timing of the introduction of new models, pricing policies should be included.

A Considered Approach

An estimate of the weight of the product concerned should be given so that a disproportionate effort is not given to relatively low weighted items. The statistical metadata system would benefit from contacts between market research organizations, retailers, manufacturers and trade associations for items for which replacement levels are high. Identify industries likely to be undergoing regular technological change. Attempt to ascertain the pace at which models change and, where possible, the timing.

Thank youThank you