Page 55 - AJWEP-22-4

Impact of cyclones on rice farming

Table 1. Descriptive statistics of the cyclone‑affected rice farmers
Variable Mean SD Min Max
Sociodemographic variables
Age (years) 48.95 12.81 20 80
Education (years of schooling) 6.24 3.84 0 18
Household size (number of members) 4.97 1.91 1 15
Earning members except the household head (number of members) 0.62 0.90 0 5
Household monthly income (BDT) 15166.25 9642.70 1000 85000
Household monthly expenditure (BDT) 14412.87 6777.22 600 50000
Farming experience (years) 21.96 12.20 2 50
Production-specific variables
Output (maunds) 35.71 40.17 5 390
Land (acres) 1.216 0.96 0.331 6.612
Labor (man-days) 15.72 13.97 2 100
Seed (kg) 39.653 42.12 6 500
Chemical fertilizer (kg) 68.302 57.43 5 380
Pesticide (kg) 0.951 1.07 0.002 8
Notes: Total respondents=400.
Abbreviations: Max: Maximum; Min: Minimum; SD: Standard deviations.

Table 2. The impact of Cyclone Amphan on rice production (estimated using Cobb‑Douglas production
function model)
Variables Model 1 (Khulna) Model 2 (Satkhira) Model 3 (Combined)
Land (acres) 0.576*** (0.137) 0.792*** (0.0596) 0.713*** (0.0703)
Labor (man-days) 0.202*** (0.0475) 0.176*** (0.0456) 0.188*** (0.0316)
Seed (kg) 0.258*** (0.0986) 0.0210 (0.0560) 0.113* (0.0582)
Chemical fertilizer (kg) 0.0423 (0.0731) 0.134*** (0.0459) 0.0807* (0.0421)
Pesticide (kg) 0.0412 (0.0298) −0.0159 (0.00973) 0.00232 (0.0113)
Cyclone (affected=1) −0.386*** (0.0625) −0.267*** (0.0357) −0.298*** (0.0256)
Constant 1.986*** (0.359) 2.559*** (0.163) 2.369*** (0.175)
Number of observations 400 400 800
R 2 0.856 0.875 0.847
Note: Regression coefficients are expressed in logarithmic form. Robust standard errors are reported in parentheses. The dependent
variable is rice output measured in maund (1 maund=40 kg). *p<0.10, **p<0.05, ***p<0.01.

the cyclone dummy variable indicates that the rice Model 3 presents pooled results from both districts.
production of affected farmers decreased by 38% Here, land, labor, seed, and fertilizer inputs all show
compared with the unaffected farmers in Khulna, which positive and significant effects on rice production.
is statistically significant at a 1% level. On average, cyclone-affected farmers experienced a
In Model 2, the coefficients of land, labor, and 30% reduction in rice output compared to unaffected
chemical fertilizer are positively and significantly farmers. The cyclone effect is more pronounced in
associated with rice production. A 1% increase in land, Khulna (Model 1) than in Satkhira (Model 2), indicating
labor, and chemical fertilizer increases rice output by a relatively greater impact in Khulna. The graphical
0.79%, 0.17%, and 0.13%, respectively. The coefficient representation of regression coefficients is provided
of the cyclone dummy indicates that affected farmers in Figure A1, and standardized regression coefficients
faced a 26% loss in rice production compared with the confirming the direction and magnitude of cyclone
unaffected farmers in Satkhira. impact are shown in Table A4.

Volume 22 Issue 4 (2025) 47 doi: 10.36922/AJWEP025100063