An Experimental Study of Influences of Performance-Related Payments on Timing of Delegated Stock Purchases

ABSTRACT

Performance-related bonuses are important tools for investment organizations to incentivize stock traders. Yet, two experiments indicate that bonuses rewarding short-term performance may lead to worse timing of purchases. The authors propose that hyperbolic time discounting makes participants set lower aspired purchase prices for short-term (decreasing percentage) bonuses than for long-term (increasing percentage) bonuses. For this reason purchases are made earlier for decreasing than increasing percentage bonuses, earlier for decreasing than random prices, and earlier for high price volatility than for low price volatility. Neither purchases at the lowest price or highest bonus are attained. Hyperbolic time discounting may account for bubbles observed in experimental double-auction markets.

KEYWORDS:

• Stock purchase
• Bonus
• Bounded rationality
• Heuristic judgement
• Hyperbolic time discounting

Acknowledgments

Thanks are due to Martin Holmen for comments, Anders Carlander, Daniel Peterson, and Amanda Sondefors for assistance in collecting and processing the data.

Funding

This research has been financially supported by grants to Center for Finance, School of Business, Economics, and Law, University of Gothenburg, from the Swedish Foundation for Strategic Environmental Research (MISTRA) for the program “Behavioral Impediments to Sustainable Investments” and from the Swedish Agency for Innovation Systems (VINNOVA).

Notes

1. In double-auction asset markets, traders submit buy or sell orders that determine asset prices. With the aim of investigating the efficiency of such markets, Smith et al. [1988] used asset market experiments in which participants buy or sell quantities of a single asset whose fundamental value (the expected stream of dividends) declines deterministically over time. A common finding is that market prices exhibit strong bubble and crash patterns. The decline of the fundamental value has been suggested to be one moderating factor (Kirchler et al. [2012], Nossair et al. [2001], Smith et al. [2000]). Supporting results were obtained by Stöckl et al. [2014], showing that a nonchanging fundamental value led to more efficient pricing than either a raising or declining fundamental value.

2. Such a scenario method is not uncommon in behavioral finance research. It differs from research on experimental markets (see footnote 1), but we know of no compelling reason why observations obtained with the scenario method would not in principle apply to experimental markets.

3. The proposition of an aspiration level is consistent with that people in making decisions are boundedly rational, that is satisficing instead of optimizing (e.g., Kahneman [2003], Payne et al. [1993], Selten [2001], Simon [1982]). Others (Hoffman et al. [2012], Kontek [2011], Koop and Johnson [2012]) have recently proposed similar roles of an aspiration level in risky decision making. An application in behavioral finance is the constructs of security, potential, and aspiration level introduced by Lopes [1987] to understand both individual differences in and situational influences on risky choice that Shefrin and Statman [2000] later implemented in their behavioral portfolio theory.

4. A common feature of heuristic judgments is that an easily accessible attribute is substituted for a less accessible target attribute (Kahneman and Frederick [2005]). In the experiments the price difference is an easily accessible attribute that participants may substitute for the less accessible bonus payout. Due to different framing of the purchase choice, the disclosed changes in the percentage bonus are still expected to influence the aspiration level.