How Much Will Climate Change Reduce Productivity in a High-Technology Supply Chain? Evidence from Silicon Wafer Manufacturing.

The frequency of hot days in much of the world is increasing. What is the impact of high temperatures on productivity? Can technology-based adaptation mitigate such effects of climate change? We provide some answers to these questions by examining how high outdoor temperatures affect a high-technology, precision manufacturing setting. Exploiting individual-level data on the quantity and quality of work done across 35,190 worker-shifts in a leading NYSE-listed silicon wafer maker in China, we evidence a negative effect of outdoor heat on productivity. The effects are large: in our preferred linear specification, an increase in wet bulb temperature of 10∘C causes a reduction in output of 8.3%. Temperature effects exist even though the manufacturer's work-spaces are indoors and protected by high-quality climate control systems. Results are not driven by extreme weather events and are robust to alternative modelling approaches. They illustrate the potential future adverse economic effects of climate change in most of the industrialised world.

Is it costly to deceive? People are adept at detecting gossipers' lies but may not reward honesty.

The possibility that gossipers may share dishonest reputational information is a key challenge to claims that gossip can shore up cooperation in social groups. It has been suggested that imposing social costs on dishonest gossipers should increase the honesty of these reputational signals. However, at present, there is little evidence of people's willingness to impose costs on dishonest gossipers; there is also little evidence of their ability to detect gossipers' lies in the first place. This paper aims to shed light on people's abilities to detect dishonest gossip and their treatment of those who share it. To do this, we report the results of two trust game studies using the strategy method (study 1) and repeated interactions in the laboratory (study 2). We show that in an environment where gossipers tell spontaneous lies people are more inclined to believe honest than dishonest gossip. We also show that people are more likely to treat favourably gossipers they believe to be honest, but that this does not always result in more favourable treatment for gossipers who were actually honest. We discuss the implications for the potential utility of social sanctions as a tool for securing honesty. This article is part of the theme issue 'The language of cooperation: reputation and honest signalling'.

Truth, Lies, and Gossip.

It is widely assumed that people will share inaccurate gossip for their own selfish purposes. This assumption, if true, presents a challenge to the growing body of work positing that gossip is a ready source of accurate reputational information and therefore is welfare improving. We tested this inaccuracy assumption by examining the frequency and form of spontaneous lies shared between gossiping members of networks playing a series of one-shot trust games (N = 320). We manipulated whether gossipers were or were not competing with each other. We showed that lies make up a sizeable minority of messages and are twice as frequent under gossiper competition. However, this had no discernible effect on trust levels. We attribute this to the findings that (a) gossip targets are insensitive to lies and (b) some lies are welfare enhancing. These findings suggest that lies need not prevent-and may help-gossip to serve reputational functions.

Religious Fragmentation, Social Identity and Conflict: Evidence from an Artefactual Field Experiment in India.

We examine the impact of religious identity and village-level religious fragmentation on behavior in Tullock contests. We report on a series of two-player Tullock contest experiments conducted on a sample of 516 Hindu and Muslim participants in rural West Bengal, India. Our treatments are the identity of the two players and the degree of religious fragmentation in the village where subjects reside. Our main finding is that the effect of social identity is small and inconsistent across the two religious groups in our study. While we find small but statistically significant results in line with our hypotheses in the Hindu sample, we find no statistically significant effects in the Muslim sample. This is in contrast to evidence from Chakravarty et al. (2016), who report significant differences in cooperation levels in prisoners' dilemma and stag hunt games, both in terms of village composition and identity. We attribute this to the fact that social identity may have a more powerful effect on cooperation than on conflict.