Global temperatures have registered the highest level in the past decade since the 1850s, and the frequency and intensity of extreme temperatures are likely to increase in the future with climate change. This outlook has increased concerns about the potential impacts of extreme weather, particularly in low- and middle-income economies (LMIEs), where the equipment to cope with environmental insults is scarce and the resources to invest in adaptive technologies are few. In the economic arena, recent studies suggest that extreme weather events increase firms’ costs and reduce local demand. There is evidence that these events diminish agricultural yields, reduce labor productivity, increase absenteeism, diminish local spending, and, when they induce adaptation, raise operational costs.
These impacts on costs and demand may create liquidity shortages for firms that may turn into solvency problems, especially for small and medium-sized enterprises (SMEs). SMEs have a more limited access to credit than large firms, and thus SMEs find it more difficult to cope with liquidity shortages. Furthermore, access to credit is more limited in LMIEs, where credit markets are shallow and institutions are less prepared to deal with informational asymmetries. All in all, this suggests that SMEs may not obtain the financing they need to cope with the negative effects of extreme weather in LMIEs, and this impossibility may lead them to default on their loans.
In a new working paper, we investigate the effects of extreme weather events on credit default and credit use of SMEs in Mexico, a middle-income economy. Finance is vital for SME growth in developing countries where these enterprises are the primary source of employment and job creation.1 To the best of our knowledge, this is the first study on the impact of extreme temperatures on credit performance of firms of any size and, therefore, the first one to analyze credit delinquency of SMEs in a middle-income economy.
We exploit a data set with loan-level information on loans extended by commercial banks to private firms in Mexico between 2010 and 2018 and follow Addoum et al. (2020) and Somanathan et al. (2021) in measuring exposure to extremes. In particular, our exposure variable is defined as the number of days in a quarter that minimum and maximum temperatures are below 3°C and above 36°C, respectively, which correspond to the bottom 5 percent and top 5 percent of the daily minimum and maximum temperature distribution in the country.
We construct quarterly credit delinquency rates at the municipality level and relate them to the number of anomalous days of extreme temperature that occurred in a given municipality and quarter. The identification strategy relies on the assumption that these extreme temperature shocks (i.e., the number of anomalous days of extreme temperature) are exogenous after controlling for seasonality and time trends specific to each municipality, as well as for national-level changes in credit delinquency rates over time.
The results show that(figure 1). For every 10 days of exposure to unusual extreme heat in a quarter, the delinquency rate of SMEs increases by 0.16 percentage point (8% of the sample mean). By contrast, none of the specifications yields a statistically significant result for the case of large firms. This finding is consistent with the idea that extreme weather events have a negative impact on loan defaults of SMEs because these firms are less equipped to deal with extreme temperatures and they find it more difficult to access further credit in times of financial stress in LMIEs.
We also investigate potential heterogeneous impacts of extreme temperatures across industries and regions. Temperature shocks affect economic outcomes through several channels, some of which are stronger for some industries than for others, and have different implications for extreme heat and extreme cold. First, weather is used as a direct input in the process of plant growth, and thus extreme heat has a particularly sizable impact on agriculture. Second, extreme heat reduces task efficiency and hours worked because it creates fatigue and cognitive impairment. However, the evidence is less conclusive for extreme cold. Finally, thermal stress creates discomfort and diminishes demand. These effects are larger for leisure and outdoor activities, such as shopping, dining, traveling, and personal services.
Consistent with these mechanisms, we find that the negative impacts of extreme heat are stronger in agriculture. Interestingly, however, in regions with a sufficiently large proportion of agricultural workers, extreme heat also has sizable effects in non-agricultural industries. These effects are concentrated in services and retail, that is, non-tradable sectors that rely heavily on local demand. The results are, then, suggestive of spillover effects originated in agriculture that expand to non-agricultural industries through reduced local spending. Hence, our outcomes suggest that. For extreme cold, there is a relatively small effect on delinquency rates in non-tradable industries, which is consistent with the possibility that consumer discomfort reduces the demand for leisure activities during cold days. It is also consistent with existing evidence that outdoor leisure activities are more sensitive to extreme cold than extreme heat.
In sum,. Policy makers in LMIEs could implement policies striving to achieve this goal. These policies could complement measures that more directly target exposure to climate risks in banks´ balance sheets.