The influence of religious beliefs on investment is interesting and important in literature. We use a large dataset with detailed information on the worship places of the five largest religious groups in China to study the relationship between local religious beliefs and individual financial decisions. We find that Buddhists, Taoists, Islamists, and Catholics are less likely to buy financial products on the financial market; Protestants tend to take risks compared to other believers. Chinese Protestants, unlike American Christians, seek more risk, while Catholics are risk-averse.

How do expectation biases causally affect households’ financial decisions? We exploit a unique setting and study the repayment decision in solar loans, in which households borrow to purchase and install solar photovoltaic (PV) systems. Electricity production – the benefit that solar panels generate – primarily depends on sunshine duration. This creates exogenous within-person across-period variation in recent signals that borrowers observe and thereby expectations of future electricity production. We find that a one-standard-deviation decrease in sunshine duration in the week right before the repayment date leads to a 20.8% increase of delinquency, even though deviated past sunshine duration does not predict that in the future. Survey evidence shows that agents make more positive forecasts of future electricity production after experiencing longer sunshine duration in the past week. We examine a battery of alternative explanations and rule out mechanisms based on liquidity constraints and wealth effects.

Does the ancient farming culture shape the risk-taking behavior of households today? Using a dataset covering over 130,000 households from a Chinese national survey, our study examines the relationship between the culture of rice cultivation and the financial behavior of modern households. We find that households in regions with a higher rate of historical rice cultivation are more likely to invest in the financial market and buy lottery, but less likely to purchase insurance. We also find that the rice area has more households with risk preferences consistent with prospect theory expectations. To account for omitted variable bias, we use average regional rainfall and downstream distance to ancient irrigation systems as instrumental variables for rice cultivation, and our results remain robust. We find that the rice effect cannot be explained by regional economic development, traditional Confucian values, or ethnic diversity. To explore potential mechanisms, we find that households in rice regions are more likely to borrow money from friends and relatives and have interest waived, and historical commercial development has also been influenced by the rice culture.

Evidence from the U.S. and Europe suggests that current risk assessment tools used by researchers and financial professionals to determine individuals’ risk tolerance and provide suitable portfolio recommendations may be flawed due to “mis”perceptions of risk. Limited research has examined the reliability of these tools as measures of relative risk tolerance for households in emerging economies like China. This study develops a multidimensional index of risk tolerance specifically tailored for Chinese households using a psychometric approach. The effectiveness of this multidimensional index in predicting individuals’ financial decisions is tested and compared to traditional unidimensional measures of risk tolerance commonly used in developed countries. The findings indicate that multidimensional measures are more consistent and significant predictors of Chinese households’ investment decisions. Additionally, the study uncovers evidence that cultural differences, related to market expectations and social networks, which are often overlooked in U.S. and European models, play a crucial role in shaping individuals' risk perceptions and investment choices in China. Robustness checks were conducted to account for potential endogeneity between risk tolerance and investment decisions. The findings provide valuable insights for researchers and financial professionals seeking to develop more accurate risk assessment tools that capture risk attitudes and perceptions in China and other developing countries. By adopting a multidimensional approach that accounts for cultural and psychosocial factors, these improved tools can enhance the precision of risk evaluation and facilitate more appropriate investment recommendations.

In the era of big data, asset pricing is influenced by various factors, which are extracted from multi-source heterogeneous information, such as high frequency market and sentiment information, low frequency firm characteristic and macroeconomic information. Especially, low frequency policy information plays a significant role in the long-term pricing in China but it is barely investigated due to its textual form. To this end, we first extract policy variables from major national development plans (“Five-Year Plans”, “Government Work Reports”, and “Monetary Policy Reports”) using Natural Language Processing (NLP) technique and Dynamic Topic Model (DTM). However, traditional models are inadequate for mixed frequency data modeling and feature extraction. Then, we propose a mixed frequency deep factor asset pricing model (MIDAS-DF) that solves the asset pricing problems under the mixed frequency data environment through mixed data sampling (MIDAS) technique and deep learning architecture. Time-varying latent factors and factor loadings can be modeled from mixed frequency data directly in a nonlinear and data-driven way. Thus, the MIDAS-DF model is able to learn the nonlinear joint-patterns hidden in multi-source heterogeneous information. Our empirical studies of 4939 stocks on the Chinese A-share market from January 2003 to July 2022 demonstrate that low frequency policy information has profound impacts on asset pricing, which anchors the long-term pricing direction, and high frequency market and sentiment information have significant influences on stock prices, which optimize the short-term pricing accuracy, they together enhance the pricing effects. Consequently, pricing effects the MIDAS-DF model outperform the five competing models on individual stocks, various test portfolios, and investment portfolios. Our research about heterogeneous information provides implications to the government and regulators for decision-support in policy-making and our investment portfolio is of great importance for investors’ financial decisions.

Utilizing a national online survey we conducted in China, we examine the impact of COVID-19 on individuals’ willingness to take risks, willingness to trust other people, and mental health measured by the Center for Epidemiological Studies-Depression (CES-D) scale. Our findings suggest that people who live in the neighborhood with a higher number of confirmed cases became more risk-averse, less likely to trust others, and more depressed. Interestingly, the effects on risk preferences and trust attitudes are statistically significant only for men, and the effects on depression are statistically significant only for women. Furthermore, the impact of COVID-19 on financial decisions, such as buying new commercial insurance and making a risky investment, is also statistically significant only for men, which is consistent with our findings on risk preferences. Attitudes towards cadres and doctors mainly drive the results on trust attitudes. The change in employment status does not drive these effects.

In this paper we use the data from “paipaidai”, an online peer-to-peer lending platform in China, to testify whether mood affects the efficiency of credit approval by individual. Refering to the studies in Psychology and Financial Economics, we employ season, temperature and weather as mood proxies, and crotrol the variables related to the quality of loan to study the credit approval behavior under different mood condition. The results suggest that the efficiency of credit approval is significantly correlated with mood—positive mood would improve the efficiency, while negative mood would reduce it. Specifically, loan examined under better mood condition (e.g. spring, comfortable temperature, and sunny days) has significantly higher probability of approval, but lower probability to default if approved; and that examined under lower mood condition shows lower probability of approval and higher probability to default if approved. This effect of mood is even stronger when a loan application to judge is more complex, atypical, or unusual. Moreover, investor sentiment, denoted by closed-end fund premiums, has the same effect on credit approval as well.