Abstract

Since Chen (2013), a fast-growing body of literature has documented abundant supporting evidence for the linguistic-savings hypothesis. Despite this influx of research, direct causal evidence is limited. In this study, we take advantage of a unique linguistic feature of the Chinese language: speakers can freely choose whether or not to use the future tense when referring to a future event. This flexibility allows us to unobtrusively manipulate the use of “will” in the description of the rewards in a standard time preference task to cleanly examine its effect on intertemporal decisions. However, our results do not lend further empirical support for the linguistic-savings hypothesis.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call