Simplifying difficult calculations: consumer choice of two‐part tariffs

Abstract

Purpose This paper aims to outline a decision process for how consumers choose among two‐part tariffs which consist of a flat fee plus a per unit charge for usage over an allowance. The paper also seeks to examine what types of decision aids help consumers choose lower cost tariffs. Design/methodology/approach The approach used is two experimental studies. Findings Most consumers do not choose a tariff by calculating an expected cost because of usage uncertainty. They instead rely on simple comparisons of the overage rate, usage allowance, and flat fee attributes. These heuristics lead to systematic biases, beyond what actual true cost justifies, for favorable comparisons on these attributes. An online calculator improved choice of the lower cost option from 65 percent to 80 percent, yet this increased to 91 percent if people were also forced to consider a range of usage levels. Practical implications Consumers struggle to choose the lowest cost tariff, especially with uncertain usage. Consumers should realize the biases in their decision shortcuts and use the presented decision aids. Firms can leverage these biases by offering larger usage allowances (often done) and smaller overage rates (often not done), or correct them with decision aids. Originality/value Much work on tariffs assumes consumers calculate a cost, but the authors question this assumption, and show that consumers instead use simple attribute comparisons to deal with uncertain usage. An understanding of the specific heuristic consumers use allows the authors to better account for past effects, predict and establish new effects, and design effective decision aids. Results indicate tariff biases largely result from information processing shortcomings.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 8 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.