Using two experiments, this paper demonstrates that expert players of sequential- move games best respond to their opponents’ backward-induction ability. In particular, I show that these experts take advantage of inexperienced opponents’ weakness in backward induction. I find this when the expert is explicitly told that her opponent is inexperienced, but also when she infers the opponent’s weakness from the opponent’s preceding performance. I demonstrate that other-regarding preferences have no role in these findings. I find that a novel model of limited foresight and uncertainty about the opponent’s foresight fits the data better than Level-k or Quantal Response models.

Divyanshu Jan, and Praneel Jain) Using online experiments, we study how charitable donations are causally affected by the nature of the task performed as part of one’s job. We find that donations are significantly higher for participants who were randomly allocated to a task designed to be ‘interesting’ compared to those participants who were randomly allocated to a task designed to be ‘tedious’, even though both tasks yielded equal earnings. We also measure the causal impact of making the nature-of-task salient before donation decisions. We find that this salience has a differential impact on the ‘interesting’ and ‘tedious’ tasks; salience reduces donations for the interesting task but increases donations for the tedious task.

Using a randomized-control-trial design, we study: (a) the level of support for a lockdown in rural Telangana, India, in September 2020, and (b) the causal impact of increasing COVID-salience on the support for a lockdown. As our salience intervention, we use a short audio clip containing commonly available information about COVID-19. The two control groups have a placebo audio clip which makes Dengue salient and no audio-clip respectively. We find high support (45%) for lockdowns in control groups, and, that a simple COVID-salience intervention causally

increases the willingness to continue lockdown by 25 percent and the reported appropriate number of days under lockdown by 33 percent. Assuming the second wave in India increased COVID-salience, our results suggest that there may be widespread support of a lockdown despite the economic consequences.

This paper examines behavior (theoretically and experimentally) in a two-stage group contest where the first stage comprises intra-group contests, followed by an inter-group contest in the second stage. Rewards accrue only to the members of the winning group in the inter-group contest, with the winners of the intra-group contest within that group receiving a greater reward. The model generates a discouragement effect, where losers from the first stage exert less effort in the second stage than winners. In contrast to previous frameworks of sequential contests, we show that a prior win may be disadvantageous, generating lower profits for first stage winners as compared to losers. This implies that incentives for participation in the first stage may not always be present. We also consider exogenous asymmetry between groups arising from a biased contest success function in the second stage. We show that although the asymmetry occurs in the second stage, the effect of the asymmetry plays out in the first stage, with the intra-group contest being more intense within the advantaged group. Experimental results find broad support for the qualitative predictions of the model. However, we find that relative overcontribution in the second stage by losers is higher than by winners of the first stage, implying that losers bear a higher burden of the group contribution than deemed strategic.

In an experimental test of a modified dictator game, I find that incentivizing a dictator to give at least a small proportion of her endowment drives non-incentivized giving to zero. This reduces overall giving relative to the standard dictator game. Thus, introducing strategic incentives for giving can be counterproductive.

Algorithms may enable efficient, optimized, and data-driven decision-making, and in fact this vision is one of main drivers of increasing adoption of algorithms for managerial and organizational decisions. However, the fact that these decisions are made by algorithms, rather than by people, may influence perceptions of the decisions that are made, regardless of the qualities of the actual decision-outcomes (Sundar and Nass, 2001). Furthermore, employees who are subjected to algorithmic control may not be satisfied if they believe they have tacit knowledge which the algorithm does not possess. They may also not understand how the algorithm works. All these factors may lead to perception of reduced and agency and cause them to distrust the algorithm. Johannsen and Zak (2021) in multiple studies using neuroscience have shown that trust leads to better productivity of employees and enhances organizational performance.

Therefore, we propose two research questions: 1) What is the impact of algorithmic control on Trust? What factors impact trust? 2) What can organizations do to enhance trust in algorithmic control?

It is important to assess the level of financial market valuation relative to fundamentals through suitable indicators. One of the widely employed indicators of market valuation is the cyclically adjusted PE ratio (CAPE). The research project intends to develop and maintain a frequently updated database of CAPE for the Indian market, as a barometer of market valuation. It is intended to provide guidance for financial market practitioners, including fund managers and traders to monitor the aggregate market valuation levels.

This paper aims to develop risk-based capital adequacy norms for the intermediaries that would provide reasonable comfort regarding loss-absorption capacity of the intermediaries and mitigate the risk of spillover of losses to non-defaulting clients.

Algorithms may enable efficient, optimized, and data-driven decision-making, and in fact this vision is one of main drivers of increasing adoption of algorithms for managerial and organizational decisions. However, the fact that these decisions are made by algorithms, rather than by people, may influence perceptions of the decisions that are made, regardless of the qualities of the actual decision-outcomes (Sundar and Nass, 2001). Furthermore, employees who are subjected to algorithmic control may not be satisfied if they believe they have tacit knowledge which the algorithm does not possess. They may also not understand how the algorithm works. All these factors may lead to perception of reduced and agency and cause them to distrust the algorithm. Johannsen and Zak (2021) in multiple studies using neuroscience have shown that trust leads to better productivity of employees and enhances organizational performance. Therefore, we propose two research questions:
1) What is the impact of algorithmic control on Trust? What factors impact trust?
2) What can organizations do to enhance trust in algorithmic control?

It is important to assess the level of financial market valuation relative to fundamentals through suitable indicators. One of the widely employed indicators of market valuation is the cyclically adjusted PE ratio (CAPE). The research project intends to develop and maintain a frequently updated database of CAPE for the Indian market, as a barometer of market valuation. It is intended to provide guidance for financial market practitioners, including fund managers and traders to monitor the aggregate market valuation levels.

This paper aims to develop risk-based capital adequacy norms for the intermediaries that would provide reasonable comfort regarding loss-absorption capacity of the intermediaries and mitigate the risk of spillover of losses to non-defaulting clients.