Climate shocks, vulnerability, resilience and livelihoods in rural Zambia

Figures & data

Table 1. Summary statistics of key variables used in the regressions.

Variable	Description	mean	SD		n
Dependent Variables
Vulnerable	If poor in all the three survey waves or fell into poverty by the third wave in 2019 (yes =1)	0.76	0.43		6531
Resilient	If never poor over the three survey waves, or escaped poverty by the third survey wave in 2019 (yes = 1)	0.24	0.43		6531
Independent variables
Climate risk/exposure variables
Drought risk (yes =1) †	Seasonal rainfall < 2SD of 30-year average rainfall	0.17	0.36		6531
Flood risk (yes =1)^†	Seasonal rainfall > 2SD of 30-year average rainfall	0.56	0.50		6531
Seasonal rainfall	November – March rainfall amount	971	170.10		6531
30 – year average rainfall	Average seasonal rainfall over 30 years	929	148.75		6531
30-year rainfall coeff. variation	30-year rainfall coefficient of variation	0.16	0.03		6531
Standard precipitation index	Current rainfall less 30-year average divided by 30-year SD	0.24	1.10		6531
Rainfall stress	Number of 20-day periods with < 40 mm of rain	0.29	0.53		6531
Other independent variables
Used CSA	Used min till, hybrid maize seed or fertilizer (yes = 1)	0.82	0.39		6531
Accessed credit (yes =1)	Accessed credit from any source	0.18	0.38		6531
Crop diversity	If Simpsons crop diversity index > median (yes =1)	0.45	0.50		6529
Asset index	Asset index from PCA	0.10	2.45		6524
femalehead	Female head of household (yes = 1)	0.23	0.42		6531
ageh	Age of household head (years)	52.01	14.06		6531
eduhead	Education of household head (years)	6.04	3.56		6531
dep_rat	(Number of members aged < 14 + =>65)/# aged 15–64	162.46	127.26		6319
mem0_5	# of members 0–5 years old	0.51	0.73		6531
mem5_10	# of members 5–10 years old	0.96	0.92		6531
mem10_15	# of members 10–15 years old	1.14	1.01		6531
mem15_64	# of members 15–64 years old	3.65	1.94		6531
mem_65	# of members 6 5 years and older	2.08	2.09		6531
farmsize	Farm size (ha) less rented-and borrowed-in land	5.19	17.64		6520
dmarket	Distance to market (km)	24.32	31.52		6531
Instrumental variables
CSA-advice	Received advice on aspects of CSA	0.90	0.41		6531
div-advice	Received advice on crop diversity	0.74	0.39		6531
Mem-loan	Member of a loan/credit group	0.13	0.34		6531

†reflect drought and flood indicators from spatial rainfall data. Results based on farmer perceptions for 2019 suggest that 63% and 12% of farm households encountered droughts and floods, respectively, on at least part of their farm.

Figure 1. Proportion of vulnerable and resilient rural households between 2012 and 2019 in Zambia. Notes: Own calculations from RALS (2012-2019).

Figure 2. Poverty transitions among rural households over three surveys: 2012, 2015, 2019. Notes: Own calculations from RALS (2012-2019). PPP and NNN signal poor and not poor in all three waves, respectively, whereas alternating P and N letters signals whether a household was poor (P) or not poor (N) in a given survey year. For example, PPN means poor in 2012 and 2015 but not in 2019.

Figure 3. Proportion of chronic, transient, and never poor rural households between 2012 and 2019 in Zambia. Notes: Own calculations from RALS (2012-2019).

Table 2. First stage results on the relevance of instrumental variables.

	Model 1: Used MT	Model 2: Access to loans	Model 3: Crop diversity
Received CSA advice at t-1	0.031**
	(2.081)
Member loan group		0.201***
		(11.288)
Received advice on crop diversity			0.068***
			(3.840)
.	.	.	.
.	.	.	.
.	.	.	.
Observations	6,302	6,303	6,303

Notes: Models 1–3 were estimated as separate probit regressions with a binary endogenous variable as the dependent variable. IVs for each model are shown in bold as the first three independent variables. Robust t-statistics in parentheses: ***p < 0.01, **p < 0.05, *p < 0.1.

Figure 4. Selected effects of climate exposure on vulnerability among smallholders in Zambia with 30-year average rainfall (left panel) and seasonal rainfall (right panel). Notes: A confidence interval for a point estimate that lies on either side of the zero shows that the estimate is statistically significant.

Figure 5. Selected effects of climate exposure on resilience among smallholders in Zambia with 30-year average rainfall (left panel) and seasonal rainfall (right panel). Notes: A confidence interval for a point estimate that lies on either side of the zero shows that the estimate is statistically significant.