Floods in Ecuador
Floods in Ecuador
Evidence from Ecuador suggests that natural disasters have important but short-lived impact on production and market conditions in developing countries that vary depending on firm differences in efficiency. From a development perspective, frequent short-lived climate shocks are a cause for concern, as they distract firms from engaging in upgrading efforts by hindering investment.
Editor’s note: For a broader synthesis of themes covered in this article, check out ourVoxDevLit on Climate Adaptation.
Given its role in escalating the frequency and severity of natural disasters, climate change has become an urgent global issue. A disaster related to weather, climate, or water hazards occurred every day on average from 1970 to 2019, killing 115 people and causing US$202 million in losses daily (WMO 2021). Climate change has also contributed to a surge in natural disasters that have disproportionately impacted developing countries over the past five decades (ibid.). In addition to the loss of human life, the destruction of production facilities and infrastructure have compromised industrial economic development efforts.
Moreover, households hit by destruction and job losses are likely to buy fewer products, affecting firms’ profitability as prices and sales volumes decrease. This demand shock is amplified by lowered demand from firms due to uncertainty over future economic conditions.
Why investigate the impact of the El Niño climate shock in Ecuador?
The El Niño natural weather phenomenon describes the unusual warming of surface waters in the eastern tropical Pacific Ocean resulting in unexpected episodes of heavy rainfall across many countries in different regions across the world. Ecuador was among the countries affected by these shocks in 2002-03, when heavy rainfall triggered landslides and floods devastated transportation and electricity infrastructure (UN-OCHA 2002). The International Disaster Database (EM-DAT) estimates that over 60,000 individuals were directly affected by the floods, with projected financial damages surpassing $26 million in 2002.
From the mid-1990s to early 2000s, Ecuador experienced a number of heavy rainfall episodes. The intensity of excess rainfalls varied, however, across the most and least rainfall-affected regions, as shown in Figure 1. While the least affected regions did not experience exceptional rainfall, the most affected experienced very high levels of rainfall. Our indicator of excess rainfall varies substantially across regions (Figure 1), with pronounced impacts on coastal regions (Figure 2). We exploit this variation in our econometric analysis.
Figure 1: Monthly rainfall in Ecuadorian provinces, 1993-2006
Monthly rainfall in Ecuadorian provinces, 1993-2006
Notes: Rainfall is in millimeters (mm). Source: World Bank (2025).
Figure 2: Average excess rainfall across Ecuadorian provinces in 2002
Average excess rainfall across Ecuadorian provinces in 2002
Average excess rainfall across Ecuadorian provinces in 2002
Measuring the effect of Ecuador’s 2002-03 El Niño shock on firms’ production decisions
In recent research (Bas and Paunov 2025), we investigate how heavy rainfalls resulting from the 2002-03 El Niño climate pattern affected firms’ production and market conditions in Ecuador. We assess market conditions by tracking firm output prices and the evolution of their markups, and capture production conditions using measures of firm marginal costs and production efficiency, measured by quantity total factor productivity (TFP-Q).
To assess the effect of this climate shock on firms’ production and price decisions, we use the 2002-03 El Niño heavy rainfall episode as an exogeneous climate shock. The unpredictable nature of this climate shock allows us to conduct a natural experiment, given that the exact timing and location of its future occurrences are unknown. We classify firms located in provinces that experienced excessive rainfall due to the 2002-03 shock relative to the historical mean higher (lower) than the median as those in the treatment (control) group.
We also conduct an event study analysis to validate our findings by testing for the absence of diverging pre-trends and duration of the shock. This analysis involves comparing firms in the treated group with those in the control group for each year of the sample period before (2000-2001) and after the shock (2002-2005). Figures 3a-e show statistically significant short-term impacts on treated firms in 2002-03, validating our empirical approach. We find evidence of increased marginal costs and decreased markups and revenue total factor productivity (TFP-R) in 2002 when the shock started, as well as a decrease in output prices and TFP-R in 2003.
Figures 3a-e: Event study results
Figure 3a
Figure 3a
Figure 3b
Figure 3b
Figure 3c
Figure 3c
Figure 3d
Figure 3d
Figure 3e
Figure 3e
Source: Bas and Paunov (2025).
The impact of climate change depends on initial firm performance
Our main findings show that affected firms’ revenue productivity (TFP-R) and markups decrease. This is due to production efficiency losses (TFP-Q) and higher marginal costs of initially less efficient firms. The impact on initially more efficient firms is explained by decreased product output prices in response to lower product demand. However, the shock did not affect market shares or survival rates of initially less efficient firms. This means that productivity distribution of Ecuador’s industry was not affected by the shock. We also find that production and market demand recovered swiftly in the immediate aftermath of the shock. The impacts of the 2002-03 rainfall shock mirror those of 1997-98.
Differentiating firms by their TFP-R, as opposed to TFP-Q, indicates that firms with better (worse) market positions can mitigate the negative impacts of the shock more (less). When investigating the heterogeneous effects of climate shocks dependent on initial firm performance, we rely on firm TFP-Q as it is a more accurate measure of firm productivity, whereas existing research uses firm TFP-R or sales. Using TFP-R instead of TFP-Q not only captures differences in firm efficiency, but also market power.
Figure 4 presents the results when using TFP-R instead of firm efficiency (TFP-Q). Our findings suggest that when we rely on TFP-R to measure the differences across less productive firms, the negative impact of the shock on production (both marginal cost and TFP-Q) and markups is lower than when we rely on firm efficiency (TFP-Q). This suggests that firms with more market power are also in a better position to attenuate the effects of the shock. One reason for this could be that these firms had more resources to deal with the shock’s production impacts. Interestingly, initially higher TFP-Q firms lowered their output prices more, resulting in higher markup reductions compared to initially higher TFP-Q firms.
Figure 4: Heterogeneous effect based on revenue versus quantity productivity
Heterogeneous effect based on revenue versus quantity productivity
Heterogeneous effect based on revenue versus quantity productivity
Notes: Each marker indicates the estimated coefficient for marginal costs, markups, and output price markers for TFP-Q and TFP-R. Statistically insignificant coefficients are omitted from the figure.
Furthermore, the impacts of the climate shock are short-lived as production recovers in the immediate aftermath. Our findings are in line with the results of Pelli et al. (2023) who also show that the negative effect of another type of climate shock, cyclones, on firm’s fixed assets and sales in India are temporary and only last one year. This is interesting from a climate policy perspective, as the costs of shocks depend crucially on their duration, requiring differentiation across types of climate and other health and geopolitical risks. Nonetheless, from a development perspective, frequent short-lived climate shocks are a cause for concern, as they distract firms from engaging in upgrading efforts by hindering investment (Verhoogen 2023).
Implications for climate and industrial policy
Natural disasters, such as the excess rainfall caused by the El Niño climate pattern, have large, complex impacts on production and market conditions in developing countries. Interestingly, while less efficient firms face rising production costs and efficiency losses, initially more efficient firms only experience adverse market conditions. Production and market shocks jointly prompt reductions in firms’ markups, a key source of firm production investments. Although firms are remarkably resilient in response to the natural disaster, the disruptive impacts to long-term development should not be underestimated, especially in the context of accelerated climate change. Future research is needed to explore the broader implications of firms' responses to climate shocks on economic stability, social equity, and environmental sustainability, ensuring that industrial development is integrated into growth strategies that address climate-related risks.
Disclaimer: This VoxDev is based on joint work with Caroline Paunov entitled "Riders on the storm: How do firms navigate production and market conditions amid El Niño?" published in the Journal of Development Economics in 2025. The views expressed in this blog do not necessarily represent the views of the OECD or its member countries.
References
Bas, M, and Paunov, C, (2025), "Riders on the storm: How do firms navigate production and market conditions amid El Niño?," Journal of Development Economics, 172, 103–374.
Pelli, M, Tschopp, J, Bezmaternykh, N, and Eklou, K, (2023), "In the eye of the storm: Firms and capital destruction in India," Journal of Urban Economics, 134, 103–529.
UN-OCHA, (2002), "Ecuador: Flood: 2002/03/08".
Verhoogen, E, (2023), "Firm-level upgrading in developing countries," Journal of Economic Literature, 61(4), 1410–1464.
World Meteorological Organization (WMO), (2021), "WMO atlas of mortality and economic losses from weather, climate, and water extremes 1970–2019".
World Bank, (2025), "Climate Change Knowledge Portal".
Ecuador event study climate shocks firm productivity Floods climate change Climate adaptation