Generally defined as temperature deviations from historical values, temperature shocks undermine many outcomes, including health, economic, social, and political outcomes, and threaten livelihoods globally (for a review, see Dell et al., 2014a). Concerning health, the literature shows that temperature shocks heighten mortality among infants, older adults and people with pre-existing health problems; amplify pre-existing medical issues; lead to poor mental health; increase the spread of diseases; and trigger other health-related issues, such as heat strokes, heat cramps, dehydration, disruption of health infrastructure, and cardiovascular and respiratory diseases (see, e.g., Andalón et al., 2016; Calleja-Agiusa et al., 2020; CDC, 2020; Hajat et al., 2010; Li et al., 2020; Sarmiento, 2023). At the same time, temperature shocks compromise economic growth by undermining several factors related to economic performance, including labour productivity, investment and output from production sectors (e.g., manufacturing, mining and agriculture) (see, e.g., Acevedo et al., 2020; Colacito et al., 2019; Dell et al., 2012; Dellink et al., 2017). Evidence shows that a 1°C rise in temperature, for example, lowers economic growth by 1.3 percentage points (Lee et al., 2020). Assuming these temperature shocks persist, an estimated output loss of about 9% is expected by 2100, particularly in developing countries (Acevedo et al., 2020). In the context of social and political outcomes, the literature demonstrates that temperature shocks initiate and facilitate conflicts (see, e.g., Bellemare, 2015; Bollfrass & Shaver, 2015; Salehyan & Gleditsch, 2006) and drive households into severe poverty (see, e.g., Azzarri & Signorelli, 2020; Baez et al., 2020).
We expand the literature on the impacts of weather shocks by exploring how temperature shocks influence household credit access.2 Much of what we know about the relationship between weather-related shocks, the credit market, and household financial conditions and decisions has been mainly focused on rainfall shocks and how these shocks influence loan repayment performance and borrowers’ investment decisions and repayment behaviours (for a review see, e.g., Adjognon et al., 2020; Newman & Tarp, 2020; Pelka et al., 2015). Far less is known about how temperature shocks influence households’ access to credit. Thus far, to the best of our knowledge, no study has specifically investigated the nexus between temperature shocks and household credit access. In this regard, we provide the first study to fill this important gap in the literature by exploring the impact of temperature shocks on household credit access in Malawi.
We are motivated to focus on household credit access by the literature that associates household credit access with several important household-related outcomes (see, e.g., Bauer, 2016; Quach, 2017; Salima et al., 2023; Sekyi, 2017). Notably, this literature suggests that greater access to credit can benefit households by improving food security, diversifying household income, assisting households in managing financial emergencies and coping with risks, encouraging entrepreneurship, facilitating investment and wealth creation, and allowing poor households to smooth consumption. By examining the effect of temperature shocks on household credit access, we improve our understanding of how these shocks affect households’ access to credit and how lenders behave when faced with shocks; provide relevant knowledge, which can influence the designing and implementation of policies that govern credit provision, promote financial inclusion, reduce financial exclusion and enhance the financial sector’s resilience and stability; and inform the development of climate change adaptation policies.
The impact of temperature shocks on household credit access is ambiguous. It may depend on various factors, including regional variation, timing and duration, household characteristics, financial sector response, and adaptation strategies. For instance, 1) temperature shocks may have a more significant adverse impact on households in regions that are more dependent on agriculture or that have weaker financial systems; 2) short-term temperature shocks may have a smaller impact on households’ ability to access credit than longer-term shocks that affect crop yields or other income sources; 3) households with lower income or wealth may have a more challenging time accessing credit in the face of temperature shocks; and 4) in response to increased credit risk associated with temperature shocks, lenders may become more risk averse, tightening credit standards, reducing their lending volumes or raising interest rates – making it harder for households to access credit. Evidence shows that when faced with crises and shocks, lenders are more likely to restrict credit to borrowers out of fear of default (see, e.g., McKillop et al., 2020; Udry, 1994). Alternatively, shocks can potentially enhance the availability of household credit as lenders seize the opportunity to serve a growing number of borrowers affected by the shocks. Households may also respond to temperature shocks by adopting various adaptation strategies, such as changing crops, which may increase or decrease their ability to repay loans and, consequently, impact their access to credit.
We use Malawian household data covering 2010, 2013 and 2016, collected from the World Bank Microdata Library, to examine the effect of temperature shocks on household credit access. Using district-level identifiers, we geographically match this household data with the weather data from the University of Delaware’s climate database. Consistent with the literature, household credit access is measured using 1) a dummy variable that takes the value of 1 if a household managed to acquire a loan in the previous 12 months and 2) an indicator that captures the number of loans a household managed to obtain in the past 12 months. In line with the literature, we measure temperature shocks using temperature bins. Estimating a panel fixed effects model and controlling for relevant fixed effects and covariates, we find that temperature shocks negatively affect household credit access. We identify heterogeneity in the impact of temperature shocks, indicating that, when exposed to these shocks, female-headed and rural households exhibit a decreased probability of accessing credit, whereas informal lenders are less likely to provide credit. Moreover, we explore economic growth, local conflict, household income and health as potential transmission channels mediating the relationship between temperature shocks and household credit access. By investigating the mediating roles of these channels, we provide valuable insights into the transmission mechanisms through which climate change impacts households and the financial sector. Our channel analysis results show that household income, district-level economic growth and local conflict are channels through which the impacts of temperature shocks on household credit access are mediated.
Malawi makes for an interesting case study for two main reasons. First, worldwide, Malawi is recognized as one of the countries most susceptible to the negative effects of climate change due to its disproportionate dependence on the agricultural sector for food, exports and employment (Warnatzsch & Reay, 2019). Specifically, the Malawian agriculture sector accounts for 32% of the Gross Domestic Product (GDP), 64.1% of the total labour force and 80% of the nation’s export revenues (FAO, 2021). However, rainfall is no longer predictable across the country, and temperature highs have risen, reaching 40˚C in many districts (Mountain Research Initiative, 2021). Concurrently, weather shock-induced natural disasters, such as droughts and floods, continuously pressure the country’s water supplies and food security (FAO, 2021). Second, Malawi’s financial sector has been transformed in the past decade. As of 2009, the sector was concentrated, small and serviced a narrow clientele, comprised mainly of urban dwellers (World Bank, 2021a). However, between 2011 and 2018, the country experienced digital transformation and increased credit access, raising the proportion of the adult population with access to financial services from 19–40% (World Bank, 2021a). In addition, the country has been receiving significant amounts of capital to boost financial services coverage and the financial sector’s performance. In 2020, Malawi received US$86 million credit from the World Bank to enhance the coverage of its financial services (World Bank, 2020). By positioning our study in Malawi, we can better understand how climate change affects vulnerable and agriculturally dependent households in developing countries.
We make important contributions to several strands of literature. Firstly, we contribute to the literature on the determinants of household credit access (see, e.g., Assogba et al., 2017; Cui et al., 2017; Duy et al., 2012; Kiplimo et al., 2015; Wongpit & Sisengnam, 2022) by demonstrating that temperature shocks are also another significant determinant. Secondly, we make contributions to the literature that has examined the effects of various types of shocks (see, e.g., Burke, Gong, et al., 2015; Colacito et al., 2019; García-Gómez, 2011; Matz et al., 2015; Minale, 2018). A large proportion of this literature has examined the impacts of shocks in price, agriculture productivity, health, income, and weather (e.g., rainfall and temperature) on outcomes, such as food security, migration, employment, health, economic growth, investment, poverty, conflict and disease prevalence. We contribute to this literature by specifically focusing on the effects of temperature shocks on household credit access.
Closely related studies have investigated the relationship between weather shocks and the credit market (see, e.g., Adjognon et al., 2020; Newman & Tarp, 2020; Pelka et al., 2015). These studies have improved our understanding of how weather shocks, particularly rainfall, influence credit risk and borrowers’ behaviours and investment decisions. We make additions to these studies by showing that temperature shocks influence economic growth, household income and local conflict and that these serve as important mechanisms through which temperature shocks transmit to household credit access.
Thirdly, we contribute significantly to the climate economy literature, which associates the understanding of climate change impacts with an improved probability of devising policies that may address the adverse consequences of extreme weather events (see, e.g., Cachon et al., 2012; Graff Zivin & Neidell, 2014; Seppanen et al., 2006). Specifically, we answer the call from Dell et al. (2014b)’s study, which concluded that there is a need to explore other outcomes that may significantly impact economic development. In response, we examine household credit access as the outcome – an important factor in the economic development equation (see, e.g., Amoo et al., 2017; Duican & Pop, 2015). In this regard, we contribute to both the literature on the economic effects of climate change and the broad climate economy literature.
The rest of the study is structured as follows: next, we explore the channels through which temperature shocks may influence household credit access. Section 3 discusses the data and variables, whereas Section 4 details the methodology. Sections 5 and 6 interpret the results and conclude, respectively.