Natural disasters disproportionally affect the poor. It is therefore often assumed that natural disasters increase income inequality. However, as Karim and Noy (2016) point out, there is little research on the impact of natural disasters on income inequality. This paper contributes with a study of Sri Lanka. We study the impact of natural disasters on income inequality in Sri Lanka at district level, as the first study of this nature. We find that natural disasters decrease income inequality among Sri Lankan households. These findings may be somewhat surprising on the face of it as one would expect natural disasters to exacerbate income inequality. However, at subsistence level, people possess little that can be lost to a natural disaster. Losses for the wealthier groups would be disproportionately greater due to natural disasters. People on a monthly wage would not see their income affected by a disaster, but small business owners would. Unskilled day labourers may find new opportunities in the reconstruction effort. Investigating the impact of Cyclone Aila in Sundarbans region in Bangladesh in 2009, Abdullah, Zander, Myers, Stacey, and Garnett (2016) establish that income inequality decreased after the cyclone. Another very recent paper by Feng, Lu, Nolen, and Wang (2016) show that household income fell by 14% due to 2008 Sichuan earthquake in China, however, income inequality did not change. Our findings are in line with the results of the aforesaid two studies on Bangladesh and China (Abdullah et al. (2016), Feng et al. (2016)). Our data allow us to decompose income sources, so that we better understand the mechanisms. The paper proceeds as follows. Section 2 presents a background discussion with related existing literature. Section 3 describes data and empirical strategy. Results are discussed in Section 4 followed by Section 5 which contains robustness checks. Section 6 sets out concluding remarks with some policy implications and also recognises the limitations of the study.

۲٫ Background discussion

In the aftermath of a natural catastrophe, it is essential that affected agents should have access to timely and sufficient finances to ensure a smooth and speedy recovery (Keerthiratne & Tol, 2017). Flow of foreign aid that follows a natural disaster plays a key role in the economic recovery process. Enterprises would recover fast when they are provided with additional capital after a natural disaster. Using a randomised experiment where randomly selected enterprises in Sri Lanka were given cash grants after the tsunami, De Mel, McKenzie, and Woodruff (2012) present evidence for this. Wealthy individuals are in a better position to meet the financial requirement through self-financing as they can use their savings for reconstruction, they are more likely to have bought insurance to cover any losses, and they have better access to loans and credit. Not only that, the rich are often better prepared for natural disasters as they can financially afford to have precautionary solutions to avoid or mitigate disaster damages. Further, the poor are more likely to have irregular income, so that every disruption, either due to the disaster directly or dealing with the aftermath, means a loss in income. As such, even within the same country, natural disasters would differently affect rich and poor individuals. Natural disasters may thus negatively affect the level of income of the poor leading to a widened income inequality in society.