소비자의 과거지출패턴이 소비지출에 대한 판단에 미치는 영향 (The Effects of Consumer Payment Patterns on Consumers’ Judgment of Past Spendings)

서지정보

· 발행기관 : 한국소비자학회
· 수록지 정보 : 소비자학연구 / 21권 / 1호 / 201 ~ 221페이지
· 저자명 : 하영원, 정성희

초록

과거 소비는 소비자의 현재와 미래의 판단과 소비에 영향을 미친다. 특히 본 연구에서는 과거에 일어난 일련의고통이나 즐거움에 대한 회상적인 평가가 개별적인 고통과 즐거움의 합으로 형성되는 것이 아니라 소비자들의 머릿속에서 재구성될 것이라고 보는 견해 (Loewenstein and Prelec 1993)에 입각하여, 일정 기간 동안의 과거 지출에 대한 소비자의 총체적인 회상적 평가는 일련의 지출이 형성하는 지출 금액 패턴의 특징(예컨대, 최고점과 마무리) 에 의해서 영향을 받는다는 것을 밝혔다.
실험 1에서는 두드러진 최고점(peak) 지출과 마무리(end) 소액지출 추가가 소비자의 총 월별 지출액 추산에 미치는 영향을 살펴보았다. 즉, 실험 1은 한 달 지출 총액이 일정함에도 불구하고 지출 패턴에서 두드러진 최고점 지출이 있는 경우가 없는 경우보다 소비자들이 그 달에 더 많이 지출하였다고 판단한다는 것을 보였고, 소액지출의추가는 그로 인해 실제 총지출 금액이 더 커졌음에도 불구하고 추가하지 않은 경우에 비해 소비자들이 그 달에 더적게 지출하였다고 판단하게 만들었다. 특히, 소액지출 추가가 과거지출에 대한 판단에 미치는 영향은 과거 지출 패턴에 두드러진 최고점 지출이 있는 경우에(두드러진 최고점 지출이 없는 경우에 비해서) 더 크게 나타났다.
실험 2에서는 과거 지출 패턴을 크게 3 가지, 즉, 두드러진 최고점 지출이 일련의 지출 순서 상 처음에 있는 경우, 중간에 있는 경우, 마지막에 있는 경우로 나누고 과거 지출에서 최고점의 시간적 순서상의 위치가 회상적인 평가에 미치는 영향을 알아보고자 하였다. 그 결과, 소비자들은 과거 지출 패턴에서 최고점지출이 처음에 있는 경우에중간이나 마지막에 있는 경우보다 상대적으로 과거지출을 더 많이 한 것으로 추산하였다. 마지막으로 이 연구의 발견들이 갖는 이론적 및 실무적 시사점을 논의하였다.

영어초록

Most real-world decisions are not temporally independent but influenced by past events. In particular, consumers’ retrospective evaluation of past payments often heavily affect consumers’present and future judgments and spendings. Regarding the retrospective judgments of past events, researchers have suggested that the consumers’ retrospective evaluation of past episodes is a constructed judgment(e.g., Loewenstein and Prelec 1993) rather than a retrieved estimate of cumulative spending. From this perspective, this article demonstrates that the consumers’ global and retrospective evaluation of past expenses is affected by different payment patterns(e.g., salient peak payment and an additional small payment at the end).
In Experiment 1, we asked participants(n = 124) to estimate their total monthly expenses after they were exposed to hypothetical list of their monthly expenses. Experiment 1 adopted a 2(payment pattern; peak vs. non-peak) x 2(end; end vs. no end) between-subjects factorial design. In the peak condition, there was a peak expense among the list of eleven expenses that participants were told to have occurred last month. In the non-peak condition, the consumer payment pattern was relatively flat compared with the peak condition. The total amount of expenses for both conditions was the same (405,000KRW). In the ‘end’ condition, a small expense for photocopies(1,700KRW) was added to the list, whereas there was no such additional expense in the ‘no end’ condition.
The results showed that the salient payment (i.e., peak) and an additional small payment (i.e., end) significantly affect consumers’ estimation of past total spending. Specifically, despite the constant past expenses, consumers in the peak/no end condition thought they spent more (M = 423,522KRW) than those in the no peak/no end condition(M = 391,997KRW).
This difference was statistically significant(F(1,60) = 7.34), p < .05). Thus, hypothesis 1(the peak effect) was supported. Furthermore, there was a significant interaction between payment pattern and end(F(1,120) = 6.91, p < .05). To illustrate, when there was a peak in the list, an additional small payment at the end triggered smaller estimation of their total spendings(M = 366,687KRW) than that in the no end condition(M = 423,522KRW).
The difference was statistically significant(p < .05). In contrast, when there was no peak in the list, the estimated amounts of monthly expenses were not statistically different between the end and the no end conditions(end condition: M = 389,944KRW, no end condition: M =391,997KRW, p > .10). Thus, hypothesis 2 (end effect) was supported.
In Experiment 2(n = 81), we tested hypothesis 3, which posited that the estimated amount of monthly expenses will be the largest when the peak is placed at the early part of the list compared with the cases when the peak is placed in the middle or in the later part of the list. There was a significant serial position effect regarding the impact of the peak expense (F(2, 78) = 2.605, p < .10). Multiple comparisons reveal that the estimated amount of monthly expenses in the early condition(M = 413,204KRW) was higher than either in the middle condition(M = 372,485KRW, p < .05) or in the later condition(M = 389,615KRW, p < .05).
In summary, this research demonstrates that the ‘peak-end’ rule proposed by Kahneman and his colleagues applies to the consumers’ perception of a series of their past spendings.
However, this research also identifies the conditions under which the peak and end effects occur. The most significant contribution of this research is that the peak and end effects were rigorously demonstrated in a consumer setting. Field studies investigating the effects of consumer payment patterns seem to be a promising domain for future research.

참고자료

· 없음