The effect of life-cycle cost disclosure on consumer behavior: evidence from a field experiment with cooling appliances

Abstract

Theory suggests that providing consumers with an estimated life-cycle cost (LCC) may make them buy more energy-efficient household appliances in cases where energy efficiency is cost effective. This article evaluates the link between the provision of LCC and consumer behavior by using an online field experiment for cooling appliances. Internet users arriving at a commercially operating price comparison website were randomly assigned to two experimental groups, and the groups were exposed to different visual stimuli. The control group received regular product price information, whereas the treatment group was offered additional information about estimated operating cost and total LCC. Consumers’ click behavior was evaluated with multiple regression controlling for several product characteristics (n = 1,969 clicks). We find that LCC disclosure reduces the mean specific energy use of chosen cooling appliances by 2.5% (p < 0.01), making it a potentially interesting approach for environmental policy regarding the market transformation toward more energy-efficient household appliances. However, LCC disclosure also decreases the number of clicks from the price comparison website to final retailers by about 23% (p < 0.01), which makes it—in the format chosen here—undesirable from a business perspective. Therefore, future research should clarify under what (if any) conditions can monetary energy cost disclosure be associated with more positive effects for price comparison websites.

Appendix

Derivation of an equivalent time horizon for indirect discounting

For the purpose of indirect discounting, we derive a reduced, equivalent time horizon from a given implicit discount rate. The default implicit discount rate we chose as a starting point equals about 18% and is based on a German survey concerning hypothetical refrigerator purchase decisions (Kuckartz and Rheingans-Heintze 2004).¹⁶

Given an implicit discount rate of 18% and the known average service life of household appliances in Germany, one can calculate a reduced time horizon, which can substitute for direct discounting. In other words, one can determine the operating cost for an ETH that must be equal to the conventionally discounted operating cost. The general condition is given as

$$ \sum\limits_{t = 1}^T {{C_t}{{\left( {1 + r} \right)}^{ - t}}} \mathop { = }\limits^! \sum\limits_{t = 1}^{\rm{ETH}} {{C_t}} $$

(7)

where C _t is the annual operating cost in year t, T is the known average service life of a given household appliance, r is the discount rate, and ETH is the equivalent time horizon. For constant C _t (as assumed here), this expression can be reduced to

$$ \sum\limits_{t = 1}^T {{{\left( {1 + r} \right)}^{ - t}}} \mathop { = }\limits^! \sum\limits_{t = 1}^{\rm{ETH}} { = {\hbox{ETH}}} $$

(8)

Given an exogenous implicit discount rate of 18% and a known average service life of 14.4 years for refrigerators, the equivalent time horizon equals about 5 years. Practically, in the experiment, these 5 years are presented to consumers as a default value without making any explicit reference to the concept of discounting.

When consumers adjust the time horizon to their personal needs, they implicitly change the discount rate. The resulting overall equation for LCC is thus given as

$$ {\hbox{LCC}} = P + {\hbox{ETH}} \times {P_{\rm{E}}} \times {C_{\rm{E}}} $$

(9)

where P is the appliance purchase price [€], ETH is the equivalent time horizon [years], P _E is the price of electricity [€/kWh], and C _E is the specific energy consumption [kWh/year].

One problem in setting up the experiment is the difference in the known service life (see Table 2) between refrigerators (14.4 years) and freezers (18 years), which would have required differentiated time horizons for each appliance type. Consumers, however, were able to switch between appliance types and may have assumed that the estimation of operating cost remains the same for all appliances. Therefore, we applied one common service life (14.4 years) and one equivalent time horizon of 5 years to all cooling appliances.