Diapositiva 1

1
Joint UNECE/ILO Meeting on Consumer Price Indices
The interpretation of the PPPs a method for
measuring the factors that affect the comparisons
and the integration with the CPI work at regional
level
Luigi Biggeri1 , Rita De Carli2 and Tiziana
Laureti31Istat and University of Firenze, Italy
2Istat, Italy 3University of Tuscia,
Italy Geneva, 8-9 May 2008
2
Joint UNECE/ILO Meeting on Consumer Price Indices
Structure of the paper

• Introduction
• Difficulties in interpreting PPPs comparability
  and representativeness
• A method for measuring the factors that affect
  the comparisons of consumer price levels
• Preliminary results from one experiment
• Concluding remarks

Geneva, 8-9 May 2008
3
Joint UNECE/ILO Meeting on Consumer Price Indices

• Introduction

Investigate the factors that can affects
comparisons of consumer price levels between two
areas
The aim of the paper
focusing on the comparisons across different
areas or regions within a country
suggest a simple method to evaluate the
importance of the different factors
Our proposal is theoretical in order to point out
the methods and the detailed data which are
necessary to interpret adequately the binary
consumer spatial indices

• The measures could be helpful
• for deciding to what extent the products included
  in the list for the computation of spatial
  indices should be comparable and/or
  representative and characteristic of the
  household consumption
• for integrating CPI and PPP work, or to balance
  comparability and representativeness

Geneva, 8-9 May 2008
4
Joint UNECE/ILO Meeting on Consumer Price Indices
2. Difficulties in interpreting PPPs
comparability and representativeness (a)

• Difficulties of the interpretation and
  criticisms to binary comparisons of PPPs

is due to the conflicting choice between
comparability and representativeness

• To clarify the problems we need for

• Clear definition of comparability, starting with
  the definition of strict comparability for
  identical products which then may be broadened

• More precise definition of representativeness of
  products included in the comparison within the
  basic headings (BH)

Geneva, 8-9 May 2008
5
Joint UNECE/ILO Meeting on Consumer Price Indices
2. Difficulties in interpreting PPPs
comparability and representativeness (b)
The products within a BH should be representative
in terms of sample design and for consumer prices
in terms of share of households expenditures
Different degree of representativeness
One or more products are characteristic in one of
the two areas of vice-versa
Different system of weights
Geneva, 8-9 May 2008
6
Joint UNECE/ILO Meeting on Consumer Price Indices
3. A method for measuring the factors that affect
the comparisons of consumer price levels (a)

• A framework for the comparison at BH level

• Only Binary comparisons for consumer price level

• The hypothesis is that all data on prices and
  weights at products level are available

• We present two different hypothetical situations

Geneva, 8-9 May 2008
7
Joint UNECE/ILO Meeting on Consumer Price Indices
3. A method for measuring the factors that affect
the comparisons of price levels (b)
the same number of products with the same
characteristics, therefore strictly comparable,
are available both in the areas j and l
Case1a (purely theoretical condition)
Case1b (a more probable situation )
Geneva, 8-9 May 2008
8
Joint UNECE/ILO Meeting on Consumer Price Indices
3. A method for measuring the factors that affect
the comparisons of price levels (c)
the systems of weights are equal
Case1a
2
Average PricesParity
the systems of weights are different, with a
different degree of representativeness in the two
areas
Case1b
3
5bis
Geneva, 8-9 May 2008
9
Joint UNECE/ILO Meeting on Consumer Price Indices
3. A method for measuring the factors that affect
the comparisons of price levels (d)

• usual situation when making binary comparisons

Overlapping area
Outer areas separate groups of products
considered in the comparisons which can be found
in area j but not in area l, and vice-versa
Group of identical products (strictly comparable)
with different systems of weights in the two
areas (j and l)
two more different situations (pag. 6 of paper)
Geneva, 8-9 May 2008
10
Joint UNECE/ILO Meeting on Consumer Price Indices
3. A method for measuring the factors that affect
the comparisons of price levels (e)
Considering the right hand side of expression
5bis and taking into account all the products
in the two areas, a new APP could be obtained by
introducing a multiplicative factor as follows
10
WE
PPE
CE characteristicity effect
The ratio between the average prices of the
characteristic products in the two areas measures
the effect due to the differences in their
consumption baskets
Geneva, 8-9 May 2008
11
Joint UNECE/ILO Meeting on Consumer Price Indices
3. A method for measuring the factors that affect
the comparisons of consumer price levels (f)

• Using the multiplicative decomposition when the
  value of the effects are greater than 1, it means
  that there is a positive influence and vice-versa
  when they are less than 1

• It is possible to obtain four decompositions

• It is possible to aggregate the APP above the BH
  level and also and also the three different
  factors

• To implement the decomposition we need to have
  all the elementary information on prices and
  weights for the groups of products with different
  comparability and representativeness

Geneva, 8-9 May 2008
12
Joint UNECE/ILO Meeting on Consumer Price Indices
4. Preliminary results from one experiment to
measure the factors that affect APP at
territorial level in Italy (a)

• At the level of areas within a country the
  consumption baskets may be more similar (more
  strictly comparable products)

• More comparable data at detailed elementary
  level are usually available for the areas within
  a country (collected for CPIs surveys)

• Istat started a research project in 2005 to
  carry out an experiment to calculate CPPs to
  compare the consumer price levels at regional
  level (De Carli, Room document)

To test the possibility of using and integrating
the statistical information currently supplied
within the CPI surveys
Geneva, 8-9 May 2008
13
Joint UNECE/ILO Meeting on Consumer Price Indices
4. Preliminary results from one experiment to
measure the factors that affect APP at
territorial level in Italy (b)

• The experiment has been conducted for 3
  expenditure divisions

Food an beverages
Clothing and fotwear
Furniture
New ad hoc surveys
CPIs surveys

• The experiment refers to 20 cities

• Having all the necessary price data we attempted
  to calculate the APPs and to decompose them in
  the three components

• Unfortunately we do not have on expenditure
  share at level of products, so the measure
  obtained are approximate

Geneva, 8-9 May 2008
14
Joint UNECE/ILO Meeting on Consumer Price Indices
4. Preliminary results and concluding remarks

• It is clear that the weight effect and the
  characteristicity effect opay an important role
  in the binary comparisons of consumer price
  levels between the different cities in Italy

• The results regarding food products show that
  for some cities the weight effect is
  substantially less than 1 (min0.700)

and the characteristicity effect is more than 1
(max 1.535)

• For other cities the weight effect is
  substantially more than 1 (max 1.309)

and the characteristicity effect less than 1 (min
0.586)

• Results obtained are very encouraging because
  they stress the importance to have this kind of
  measures to make correct decisions in the
  implementation of CPPs

• Further researches are necessary

Geneva, 8-9 May 2008