Behavioural Economics: The Inferior-Good Effect

Abstract

These experiments used 15 domestic hens to investigate the inferior-good effect, a decrease in consumption of a commodity as income increases. Experiment 1 investigated plain and salted wheat to serve as superior goods (opposite to inferior goods) and inferior goods. Hens consumed mostly plain wheat when given ad-libitum access so it was the intended superior good and salted wheat was the intended inferior good. In the next experiments, 3 s of plain wheat and 10 s of salted wheat were available for single responses on 2 keys during discrete trials. Income changed by changing the inter-trial interval (ITI) of fixed-length sessions or the total number of trials. Experiment 2 partially replicated the Silberberg et al. (1987) procedure, using the ITI income analogue for 6 hens. When income increased, 4 hens sometimes responded less on the salted-wheat key (demonstrating the inferior-good effect), 2 of these hens and a 5th hen sometimes responded more on both keys but proportionally more on the plain-wheat key (termed here as a relative inferior-good effect). Experiment 3a partially replicated the Hastjarjo et al. (1990a) procedure using the total-trials analogue for 7 other hens. The inferior-good effect occurred across some conditions for 2 hens while other hens tended to respond on the plain-wheat key, suggesting lack of contingency contact. When a 60-s ITI was added in Experiment 3b, variability increased for most hens, but only 1 hen showed the effect, 1 of the hens that did so in Experiment 3b. These 6 hens' (1 died) 80% bodyweights were re-assessed in Experiment 4 and hens were below 80% during Experiments 3a and 3b, suggesting that the lack of the inferior-good effect was not due to some hens being at high weights. The ITI analogue was used for these 6 hens in Experiment 5 and the effect occurred for 2 hens. Experiment 6 added forced-choice trials to the total-trials analogue (with 60-s ITI) to guarantee contingency contact. The inferior- and/or relative-inferior-good effect occurred for 3 hens. Across Experiments 2 through 6, body weights were usually heavier in high-income conditions and a within-session pattern of early-salted-late-plain responding occasionally occurred. Crop capacities of 5 Experiment-2 hens and a new hen (1 died) were assessed in Experiment 7 and there was no relation between this measure and inferior-or relative-inferior-good effects. Experiments 8 and 9 examined effects of pre- and post-feed in low-income conditions using the ITI analogue. When hens were pre-fed, responding for 5 of 7 hens resembled responding in high-income conditions of Experiment 2 with more plain-wheat responding and similar or less salted-wheat responding in some conditions (behaviour similar to the inferior- and relative-inferior-good effects, but without the income change). A similar pattern was found for 4 of 5 hens when hens were post-fed in Experiment 9, suggesting that food in the digestive tract may have played a role, and perhaps not the income manipulations themselves, where it (or other component of body weight) may have abolished quantity (i.e., the intended-inferior-good) as a reinforcer. Although these experiments occasionally demonstrated inferior- and relative-inferior-good effects, but less convincingly than published studies, the effects of income may have been non-specific. The usefulness of the inferior-good concept and other income-related economic concepts are thus challenged.

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 — 0 citations exist upstream.

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