Abstract:<p>The purpose of this study is to explore how combined effect of feedback frequency of investment performance and manager's affective commitment on reduction of the sunk cost effect. To this end, we designed an experimental questionnaire to collect data from production managers of electronics manufacturing companies listed on Taiwanese stock markets and used a hierarchical regression model to examine the relationships among variables. The results from 336 samples showed that just considering performance feedback frequency or affective commitment does not necessarily reduce the sunk cost effect. Only high feedback frequency jointed with high affective commitment can suppress the willingness of manager to continue a disadvantageous investment project.</p>
What problem does this paper attempt to address?
### Problems the paper attempts to solve
This research aims to explore the combined influence of investment performance feedback frequency and managers' affective commitment on reducing the sunk - cost effect. Specifically, the author designed an experimental questionnaire, collected data from production managers in electronic manufacturing companies listed on the Taiwan stock market, and used a hierarchical regression model to test the relationships between variables. The research results show that considering only performance feedback frequency or affective commitment does not necessarily reduce the sunk - cost effect. Only when high - feedback frequency is combined with high affective commitment can managers' willingness to continue unfavorable investment projects be suppressed.
### Research background
Sunk cost refers to the cost that has occurred and cannot be recovered in the project. Although sunk cost should not affect current and subsequent decisions, many empirical studies have shown that sunk cost does indeed affect decision - makers' investment decisions. When corporate managers are unwilling to abandon unfavorable investment projects and continue to invest because of the costs already invested (i.e., sunk costs), this behavior is called the sunk - cost effect. Previous research has mainly discussed the potential causes of the sunk - cost effect, such as cognitive dissonance, and possible moderating factors, such as project completion effect, information asymmetry, and decision - type. However, there is a lack of empirical research exploring the reasons for reducing the sunk - cost effect.
### Research methods
1. **Participants and questionnaires**:
- The research subjects are production managers in electronic manufacturing companies on the Taiwan stock market, including industries such as semiconductors, optoelectronics, communication networks, computer peripherals, and electronic components.
- A total of 920 companies were selected, and the production managers of these companies were invited to participate in the survey.
- The experimental questionnaire designed four combinations of sunk - cost levels and four performance feedback frequencies.
- A total of 341 responses were received. After excluding 5 incomplete answers, the number of valid responses was 336, and the effective recovery rate was 36.5%.
2. **Experimental scenarios and procedures**:
- Managers are in charge of an investment project involving the production of innovative products, and personal rewards are closely related to project performance.
- The project budget is $100 million and needs to be completed within 18 months.
- Managers were informed that competitors had launched similar products with reasonable prices and more functions in the market, and then they decided their willingness to continue investing in the project according to the given information.
3. **Measurement and analysis**:
- **Sunk cost**: Represented by the completion ratio, divided into four levels: 15%, 40%, 65%, and 90%.
- **Performance feedback frequency**: Divided into four intervals: 3 months, 6 months, 9 months, and 12 months.
- **Willingness to continue investing**: Represented by a scale from 0% to 100%, where 0% means "absolutely unwilling to continue" and 100% means "absolutely willing to continue".
- **Affective commitment**: Measured by eight items of Allen and Meyer (1990)'s affective commitment scale, using a 7 - point Likert scale.
- **Control variable**: Age is used as a control variable.
### Results
1. **Manipulation checks**:
- The effective manipulation of sunk cost and performance feedback frequency was verified by comparing the mean differences in the willingness to continue investing.
- The effects of sunk - cost levels and performance feedback frequency on the willingness to continue investing were both significant (F - values were 8.504 and 19.957 respectively, p < 0.01).
2. **Descriptive statistics and correlation analysis**:
- The mean value of managers' willingness to continue investing was 0.405, indicating that they considered the investment project less favorable.
- There was a significant positive correlation between sunk cost and the willingness to continue investing (r = 0.371, p < 0.01), confirming the sunk - cost effect in the experimental setup.
- Performance feedback frequency and affective commitment had no direct relationship with the sunk - cost effect.
3. **Hypothesis testing**:
- A hierarchical regression model was used to test the interaction effects of sunk cost, performance feedback frequency, and affective commitment on the willingness to continue investing.
- The results showed that the interaction terms of performance feedback frequency and affective commitment with sunk cost were significant (β8 = 0.008, p < 0.01), supporting Hypothesis 1.
- Under the 3 - month feedback interval, the impact of affective commitment on reducing the sunk - cost effect was the greatest, supporting Hypothesis 2.
### Conclusions
When making investment decisions, corporate managers are often affected by the sunk - cost effect and are unwilling to abandon unfavorable investment projects, resulting in huge losses. This research found that only when there is a high performance feedback frequency...