27.10.2012 Views

Analysis of Sales Promotion Effects on Household Purchase Behavior

Analysis of Sales Promotion Effects on Household Purchase Behavior

Analysis of Sales Promotion Effects on Household Purchase Behavior

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

elasticities can be attributed to three sets <str<strong>on</strong>g>of</str<strong>on</strong>g> exogenous variables: category factors, brand<br />

factors, and c<strong>on</strong>sumer factors. As Gupta (1988), they c<strong>on</strong>cluded that switching (i.e., sec<strong>on</strong>dary<br />

demand) is the most important effect <str<strong>on</strong>g>of</str<strong>on</strong>g> a promoti<strong>on</strong>. However, that effect is not as dominant<br />

as reported by previously reported. <str<strong>on</strong>g>Promoti<strong>on</strong></str<strong>on</strong>g>s can also have a significant effect <strong>on</strong> primary<br />

demand for a product (i.e., purchase time and quantity choice). Furthermore, the magnitude <str<strong>on</strong>g>of</str<strong>on</strong>g><br />

primary and sec<strong>on</strong>dary demand effects were found to vary substantially across brands and<br />

categories. The choice elasticity varied from 49% for butter to 94% for margarine. The time<br />

elasticity varied from 1% for liquid detergents to 42% for butter. The quantity elasticity varied<br />

from almost 0% for margarine to 45% for c<str<strong>on</strong>g>of</str<strong>on</strong>g>fee. The overall average decomposes elasticities<br />

into 75/11/14 percent for respectively brand/time/quantity. Up to 70% <str<strong>on</strong>g>of</str<strong>on</strong>g> this variance was<br />

explained by the category-, brand-, and c<strong>on</strong>sumer-specific factors, in this order <str<strong>on</strong>g>of</str<strong>on</strong>g> importance.<br />

All refrigerated products had much higher timing effects than quantity effects. All storable<br />

products showed the opposite patterns. Recall that Gupta (1988) obtained different results for<br />

the storable product category c<str<strong>on</strong>g>of</str<strong>on</strong>g>fee. Bell et al. (1999) attributed this to two factors. First, they<br />

use newer and different data. Sec<strong>on</strong>d, while Gupta’s model addresses the ‘when’ questi<strong>on</strong> <str<strong>on</strong>g>of</str<strong>on</strong>g><br />

purchase timing, Bell et al. focus <strong>on</strong> the ‘whether’ questi<strong>on</strong>. No differences were found in<br />

brand choice elasticities related to storability. The overall elasticity decompositi<strong>on</strong><br />

distinguishing between storable and n<strong>on</strong>storable products resulted in 75/3/22 for storable<br />

products versus 75/17/8 for n<strong>on</strong>storable products.<br />

But, as noted and researched by Van Heerde et al. (2001, 2002), there is a big<br />

difference between the promoti<strong>on</strong>al bump decompositi<strong>on</strong> depending <strong>on</strong> whether this is<br />

derived using elasticities or unit-sales effects. Researchers decomposing the sales promoti<strong>on</strong><br />

elasticity into brand switching, purchase quantity, and purchase timing (e.g., Gupta 1988,<br />

Chiang 1991, Chinatagunta 1993, Bucklin et al. 1998, Bell et al. 1999) c<strong>on</strong>cluded <strong>on</strong> average<br />

that 74% is due to brand switching (sec<strong>on</strong>dary demand effects) and the remainder is due to<br />

timing accelerati<strong>on</strong> and quantity increases (primary demand effects). Van Heerde et al. (2001,<br />

2002) argued that the decompositi<strong>on</strong> <str<strong>on</strong>g>of</str<strong>on</strong>g> unit sales effects is theoretically and managerially<br />

more relevant than the decompositi<strong>on</strong> <str<strong>on</strong>g>of</str<strong>on</strong>g> elasticities. The former decompositi<strong>on</strong> c<strong>on</strong>siders<br />

promoti<strong>on</strong>al sales effects in terms <str<strong>on</strong>g>of</str<strong>on</strong>g> comparable units (unit sales), whereas the latter is<br />

c<strong>on</strong>structed in terms <str<strong>on</strong>g>of</str<strong>on</strong>g> percentage changes <str<strong>on</strong>g>of</str<strong>on</strong>g> n<strong>on</strong>-comparable units (probabilities and<br />

purchase quantities). Van Heerde et al. (2001, 2002) transformed the elasticity-based results<br />

79

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!