Cross-Price Elasticities of Demand Across 114 Countries
by Anita Regmi and James L. Seale, Jr.
Technical Bulletin No. (TB-1925) 89 pp, March 2010
The extent to which price fluctuations in one good may affect
the demand for other goods (that is, the cross-price elasticities)
varies across goods and countries. This report estimates
cross-price elasticities across 114 countries for 9 major
consumption categories: food, beverage, and tobacco; clothing and
footwear; education; gross rent, fuel, and power; house furnishings
and operations; medical care; recreation; transport and
communications; and "other" items. In addition, the analysis
provides cross-price elasticities for a two-good demand system
involving food and nonfood items. The estimates are based on 1996
price data from the International Comparison Program, a
multinational statistical project that collects price data that can
be compared internationally for a basket of goods and services. The
elasticities presented here are the only consistent, cross-country,
cross-price estimates available for such a large number of
countries and consumption categories.
What Is the Issue?
Price increases in goods and services may induce consumers to
substitute cheaper products for more expensive-but perhaps higher
quality-products that they feel they can no longer afford. Price
increases also reduce the real income of consumers, leaving less
disposable income for other goods and services. At the global
level, price increases may reverse gains made by poor nations
toward improved nutrition and welfare for their populations. This
study's estimates of consumer response to rising and falling prices
across 114 countries can help policymakers assess future needs for
goods and services, as well as for related transport and
infrastructure. Awareness of demand and consumption trends across
countries, essential for predicting shifts in demand for different
products, can be a valuable tool for those in the agricultural,
manufacturing, education, health, and energy sectors.
What Did the Study Find?
For each category of goods, the study estimated 2 sets of
cross-price elasticities for 114 countries:
(1) the compensated Slutsky elasticity, calculated assuming that
the real level of income stays constant, compensated by an amount
equivalent to the price rise of the good; and (2) the uncompensated
Cournot elasticity, calculated assuming that the existing level of
income stays constant, with no money added to cover the good's
Two-Good (Food-Nonfood) Elasticity
In the two-good (food-nonfood) model, an increase in the price
of the nonfood good, when accompanied with a compensating increase
in income (the Slutsky elasticity), prompts increased food demand
in all countries, but the middle-income countries experience the
greatest increase in food demand.
By comparison, an increase in the price of a nonfood good, when
not accompanied by a compensating increase in income (the Cournot
elasticity), leads to increased food demand in the poorest
countries and declining food demand in wealthier countries.
Nine-Good Elasticity Comparisons
In the nine-good model, when price increases are offset by
equivalent income increases (Slutsky elasticities), the model
- A price increase for one good triggers increased demand for the
other eight goods; that is, the Slutsky cross-price elasticities
are positive. But the demand increases for the other goods are not
equal. Demand for the more luxurious of the remaining eight goods,
such as recreation, increases more than demand for a necessity,
such as food or clothing.
- When the price of a necessity (food or clothing) rises, the
increases in demand for the other eight goods are greatest in the
low-income countries and smallest in the high-income countries.
That is, the Slutsky elasticities decrease as the income level
trends upward across the 114 countries.
- When the price of a non-necessity rises, increases in demand
for the other eight goods (except for food) are smaller in the
low-income countries than in the high-income countries; that is,
the Slutsky cross-price elasticities are lower in the low-income
countries and higher in the high-income countries.
In the nine-good model, when there is no increase in income to
offset the price increases (the Cournot elasticities), the model
- When a price increases for a necessity, the demand for all the
other goods declines; that is, the Cournot cross-price elasticities
for food and clothing are negative. That is true for all countries:
low-income, middle-income, and highincome.
- When a price increases for a non-necessity, the demand response
varies, based on the country's income level and the good in
question. For example:
- In all countries, when the price of recreation increases, the
demand for other goods rises; that is, the Cournot cross-price
elasticities are positive for recreation.
- On the other hand, when the price of transport and
communications rises, the demand for other goods goes up in
low-income countries, but declines in middle- and high-income
countries. That is, the Cournot cross-price elasticity for
transport and communications is positive in low-income countries
and negative in high-income countries.
How Was the Study
This study builds on the models and findings from an earlier ERS
report, International Evidence on Food Consumption
Patterns (TB-1904, 2003), to derive a methodology that can be
used to estimate cross-price relationships. That study estimated
income and own-price elasticities for a 9-good system (that is, for
9 broad consumer consumption categories) across 114 countries.
Using the parameters and estimated income and own-price
elasticities from the earlier ERS report, the present study
describes a framework for estimating cross-price elasticities. The
authors then calculate the compensated and uncompensated
cross-price elasticities for the same 9 consumer consumption
categories as the earlier report, as well as for a 2-good,
food-nonfood pairing, across the same 114 countries. Estimates from
the earlier study have been widely used as input to economic models
such as USDA's Baseline model. The cross-country elasticity
estimates from the present study should also provide valuable input
for other economic models designed to forecast consumer demand when
prices change. The methodology for obtaining the elasticities will
aid future ERS efforts at estimating cross-price elasticities.