Page 76 - Microeconomics, Fourth Edition

c02demandandsupplyanalysis.qxd 6/21/10 8:50 AM Page 50

50 CHAPTER 2 DEMAND AND SUPPLY ANALYSIS
lower-priced brands whose prices did not change. Thus, even if demand is inelastic at
the market level, it can be highly elastic at the individual brand level.
The distinction between market-level and brand-level elasticities reflects the
impact of substitution possibilities on the degree to which consumers are sensitive
to price. In the case of cigarettes, for example, a typical smoker needs cigarettes be-
cause there are no good alternatives. But that smoker doesn’t necessarily need Salem
cigarettes because, when the price of Salem goes up, switching to another brand will
provide more or less the same degree of satisfaction.
What determines whether a firm should use market-level or brand-level elastic-
ity in assessing the effect of a price change? The answer depends on what the firm
expects its competitors to do. If a firm expects its rivals to quickly match its price
change, then the market-level elasticity will provide the appropriate measure of how
the demand for the firm’s product is likely to change with price. If, by contrast, a firm
expects its rivals not to match its price change (or to do so only after a long time lag),
then the brand-level elasticity is appropriate.

APPLICA TION 2.3

How People Buy Cars: The Sentra), which are the most crowded. By contrast, de-
mands for cars in the luxury segment (Lexus LS400,
Importance of Brands
BMW 735i) are somewhat less price elastic because
there are fewer substitutes for them.
Using modern statistical techniques, Steven Berry,
James Levinsohn, and Ariel Pakes estimated price elas-
ticities of demand for numerous makes of automo- TABLE 2.3 Estimates of Price Elasticities of
biles. 12 Table 2.3 shows some of their estimates. These Demand for Selected Makes of Automobiles, 1990
estimates illustrate that demands for individual mod-
Model Price Estimated Q,P
els of automobiles are highly elastic (between 3.5
and 6.5). By contrast, estimates of the market-level Mazda 323 $ 5,039 6.358
price elasticity of demand for automobiles generally Nissan Sentra $ 5,661 6.528
fall between 0.8 and 1.5. 13 This highlights the dis- Ford Escort $ 5,663 6.031
tinction between brand-level price elasticity of de- Chevrolet Cavalier $ 5,797 6.433
mand and market-level price elasticity of demand. Honda Accord $ 9,292 4.798
Brand-level price elasticities of demand are more Ford Taurus $ 9,671 4.220
negative than market-level price elasticities of demand Buick Century $ 10,138 6.755
because consumers have greater substitution possibili- Nissan Maxima $ 13,695 4.845
$18,944
4.134
Acura Legend
ties when only one firm raises its price. This suggests Lincoln Town Car $ 21,412 4.320
that the most negative brand-level elasticities for auto- Cadillac Seville $24,544 3.973
mobiles should be in those market segments in which Lexus LS400 $27,544 3.085
consumers have the greatest substitution possibilities. BMW 735i $37,490 3.515
The data in Table 2.3 bear this out. The most elastic de- Source: Table V in S. Berry, J. Levinsohn, and A. Pakes,
mands are generally for automobiles in the compact “Automobile Prices in Market Equilibrium,” Econometrica,
and subcompact market segments (Mazda 323, Nissan 63 (July 1995): 841–890.

12 S. Berry, J. Levinsohn, and A. Pakes, “Automobile Prices in Market Equilibrium,” Econometrica, 63
( July 1995): 841–890.
13 See, for example, McCarthy, Patrick, “Market Price and Income Elasticities of New Vehicle Demands,”
Review of Economics and Statistics, 78 (August 1996): 543–547.