Climatic Shocks and Child Human Capital: Evidence from Ethiopia

Abstract

We examine the impact of four different climatic shocks as perceived by households and community representatives on child learning and health outcomes in Ethiopia; one of the poorest countries in Sub-Saharan Africa. Two waves of household panel data for the years 2006 and 2009 are used and data is collected from both urban and rural areas of Ethiopia. For child learning outcomes we use enrollment, Peabody picture vocabulary and math test scores and for health outcomes we use body mass index z-scores and height for age z-scores.

Keywords

Climatic shocks child learning child health and nutritional status Sub-Saharan Africa aid

Introduction

In recent years considerable attention has been devoted to the investigation of the causes and effects of various dimensions of climatic shocks. The first decade of the 21st century was the warmest decade recorded since modern measurements began in around 1850 (World Meteorological Organization, 2013). During this decade most of the world experienced above average temperatures. The overall increase in temperature suggests that the frequency of natural disasters will increase, and accordingly, expenditure on disaster management will increase. A corresponding decrease is likely to be observed in funding for household welfare. Moreover, investment in child nutrition and schooling will become more critical especially in underdeveloped societies. Besides observing the irregularities in temperature, the study of extreme events and mapping the behavior of individuals, households and institutions in different regions is important for controlling socio-economic vulnerability. Droughts, floods, storms and other events have the potential to disrupt people’s lives, leading to losses of income, assets and opportunities.

People all over Africa are vulnerable to droughts and floods and many depend on rain-fed agriculture as their primary means of subsistence. These people also often live in degraded areas susceptible to rainfall variation (cleared of trees and vegetation). Poor harvests due to rainfall variability have often led to famines and have badly affected African economies. According to the International Strategy for Disaster Reduction (ISDR) (2004: 149), the African continent is exposed to disaster risk from various natural causes, particularly those arising from hydro meteorological hazards. The most widespread droughts occurred in 1973 (killing 100,000 people in Sahel) and 1984. Later, in 1992, Southern African countries experienced severe food shortage due to extreme drought (Gommes and Petrassi, 1996). Heavy flooding in 1998 in East Africa badly disrupted its economies; in Uganda alone more than 10,000 people were affected (Environment Report on Uganda, 1998). Madagascar was hit by a cyclone in 2002, and 70% of the crops were destroyed along with several deaths and many injuries (Center for International Disaster Information, 2002). Thus, East Africa has been affected by many natural disasters in recent years and Ethiopia is no exception – in Ethiopia almost 8 million people are in danger due to extreme climate irregularities.

Periodic shocks such as droughts and heavy rains can cause immediate crises and add to the chronic difficulties linked to high poverty levels and dependence on rain-fed agriculture. As a result, people (particularly children and women) with poor nutritional status are at higher risk of death and disease when negative shocks occur (UNICEF, 2011). Children are highly vulnerable to disasters, in part because of their particular stage of physiological and social development (Seballos et al., 2011). Investments in children’s human capital can take different forms, including investment in biological human capital (for example, health and nutritional status) and intellectual human capital (schooling and cognitive skills) (Behrman et al., 2009; Cengage, 2005) – all of which may be affected by climatic shocks.

The consequences of climatic shocks are likely to be more severe in underdeveloped societies compared to developed ones due to poorer coping strategies. Climatic shocks may have long-term and more severe consequences compared to idiosyncratic shocks because buffer mechanisms like social networks are not very effective during such shocks due to their limited scale. The coping strategies may be dependent on foreign intervention in terms of aid, help from governments and help from NGOs. In this study we restrict ourselves to climatic shocks and explore whether foreign interventions, governmental aid, and cash received from any source can help buffer such climatic shocks.

The impact of climatic shocks on child human capital is ambiguous. Ferreira and Schady (2009), for example, note that the effects of aggregate negative economic shocks on investment in schooling are theoretically ambiguous because of a tension between income and substitution effects. If capital markets are imperfect for human capital investments and smoothing over time is costly, the income effect is likely to reduce child schooling. However, the price effect through reduced returns on current child employment via fewer hours worked or less income per hour worked means that the opportunity cost of going to school tends to decline. Most empirical studies find that negative aggregate economic shocks have adverse effects on child schooling, suggesting that income effects dominate (Escobal et al., 2005; Fallon and Lucas, 2002). Duryea and Arends-Kuenning (2003) analyze urban Brazilian children aged 14–16 years and find that negative income shocks lead to an increase in dropouts from school and greater entry into the labor market. Jensen (2000) finds that school enrollment in Cote d’Ivoire declined by between one-third and one-half due to unfavorable weather shocks. Similar results are also reported by Jacoby and Skoufias (1997) for households in India, Sawada and Lokshin (1999) for Pakistan and Beegle et al. (2006) for Tanzania. Flug et al. (1998) analyze cross-country panel data from 88 countries for the period of 1970–1992 and find that income and employment volatility had significant negative effects on school enrollment in low-income countries.

The connection has already been established in literature (Burgess et al., 2011; Butt et al., 2005; Nelson, 2009; Schlenker and Lobell, 2010), but also, for health both substitution and income effect work (Ferreira and Schady, 2009). A strong relationship between birth year, rainfall and adult height in Indonesia suggests that nutrition in infancy varies with early-life rainfall, i.e. climatic conditions (Maccini and Yang, 2009). Gajigo and Schwab (2012) contribute to the debate on the ability of households to offset income shocks by showing how seasonal variations in child health relate to seasonal fluctuations in agricultural income and maternal nutrition in Gambia. Similarly, the relationship between climate change and child height is investigated in Mexico by Skoufias and Vinha (2012). Some studies in developed countries (see for example Dehejia and Lleras-Muney, 2004; Ruhm, 2003, 2005) argue in favor of a counter-cyclical effect of income on health.

The aim of this paper is to investigate the impact of climatic shocks on child school enrollment and children’s abilities and learning processes as they enter adolescence. The poverty-ridden country of Ethiopia is the focus of this study. Ethiopia is a populous sub-Saharan African country with a population of over 90 million. Ethiopia is one of the poorest countries of the world. The average Gross National Income (GNI) per capita is only US$971, which is far below the average of $1966 for sub-Saharan African countries (UNDP, 2011). Ethiopia has a low human development index of 0.383, and ranks 174 out of 187 countries on UNDP’s Human Development Index (2011). Almost 40% of Ethiopia’s rural population is living in poverty and about 29% of the population is living in extreme poverty with an income of less than US$1 per capita per day. Ethiopia’s economy is based mainly on agriculture, including crop and livestock production, which contributes 45% to the Gross Domestic Product (GDP) of the country, more than 80% to employment and over 90% to the foreign exchange earnings of the country (Mininistry of Agriculture, 2010).The Ethiopian economy, particularly the large agricultural sector, is extremely vulnerable to external shocks like climatic changes, global price fluctuations influencing exports and imports and other external factors (Shitarek, 2012). Rainfall and temperature patterns vary widely because of Ethiopia’s location in the tropics and its diverse topography. Ethiopian children suffer unacceptably high rates of chronic malnutrition and poor life expectancy, and the country’s large populations continue to live at the brink of destitution and calamity (De Waal et al., 2006).

In this study we exploit the Young Lives Older Cohort panel data of Ethiopia for round 2 (2006) and round 3 (2009) to study the impact of different climatic shocks on child human capital for adolescents. We study the impact of different climatic shocks separately because of their different natures; some shocks have long-term effects while others have effects in the short-term. Though all climatic shocks affect crop production, floods can destroy households, assets and infrastructure as well, which can further inhibit income-generating activities (Garbero and Muttarak, 2012). After floods, land becomes more fertile and there is a chance of good crops in subsequent seasons, but droughts have a different and more adverse long-term impact on soil. Therefore, we study the impacts of four climatic shocks based on households’ reported perceptions: (1) drought, (2) floods, (3) hailstorms/land erosion, (4) crop failure and the death of livestock due to disease, epidemics and pesticides. Since all these shocks are based on individuals’ perceptions, the reports of these shocks may be based on their coping strategy. Thus, we also include another measure of these shocks, “community-level shocks,” based on the response of a designated community representative about whether the community experienced these climatic shocks.

The impacts of the shocks on child human capital may differ for different human capital outcomes. The opportunity cost of time in the labor force, for example, may be more important for investments in learning through attending school than for, say, nutritional status. We consider five different outcomes: three outcomes for learning and two outcomes to measure health and nutritional status. The outcome variables for learning are: (1) child school enrollment, (2) Peabody picture vocabulary test (PPVT) scores and (3) mathematics test scores. For health and nutritional status we use body mass index and height-to-age z-scores. Thus, the objective of the study is to examine the impact of four different types of climatic shocks (droughts, floods, storms/land erosion and crop failure) on child learning and health outcomes. We also aim to elaborate the role of institutional help during crisis and the role of household characteristics in buffering these climatic shocks.

Data

We use Young Lives data for the older cohort from Ethiopia. Our panel consists of two rounds: round 2 (2006) when the children were 11–12 years old and round 3 (2009) when the children were 14–15 years old. The data are collected from both urban and rural areas in five regions: Addis Ababa, Amhara, Oromia, SNNPR and Tigray (see Figure 1). Table 1 gives summary statistics for the variables that we use in the analysis.

Figure 1.

Regions of Ethiopia investigated in the study (Source: https://maps.google.com/maps/).

Table 1.

Summary statistics.

Variable	Definition of Variables	2006					2009
		Obs	Mean	Std. dev.	Min	Max	Obs	Mean	Std. dev.	Min	Max
Enrolled=1	If enrolled in school	953	0.97	0.16	0	1	972	0.89	0.31	0	1
PPVT	Peabody Picture Vocabolary Test	439	300	50	162.75	401.4	443	299.98	15.1	254.1	355.0
Math test	Math Test	793	300	49.93	164.15	441.6	802	299.93	14.9	272.5	343.5
ZBMI	Z-Score Body Mass Index	978	−1.636	1.12	−6.24	3.29	969	−1.72	1.18	−5.61	1.68
ZHFA	Z-Score Height for Age	978	−1.394	1.27	−7.15	2.44	969	−1.37	1.29	−6.02	2.45
Age (months)	Age in months	979	145.24	3.81	135.84	155.4	938	180.36	3.58	173.2	192.2
Gender	=1 if male and zero otherwise	979	1.48	0.50	1	2	938	1.48	0.50	1	2
Wealth index	Wealth Index	979	0.29	0.17	0	0.84	938	0.348	0.16	0	0.86
Household size		979	6.50	2.05	2	13	938	6.35	2.12	1	13
Mother’s education (primary)	=1 if primary and zero otherwise	979	0.23	0.42	0	1	974	0.18	0.38	0	1
Mother’s education (secondary)	=1 if secondary and zero otherwise	979	0.25	0.43	0	1	974	0.41	0.49	0	1
Father’s education (primary)	=1 if primary and zero otherwise	979	0.28	0.45	0	1	974	0.18	0.38	0	1
Father’s education (secondary)	=1 if secondary and zero otherwise	979	0.38	0.48	0	1	974	0.67	0.46	0	1
Drought	=1 if experinaced drought during last four years	979	0.31	0.46	0	1	938	0.37	0.48	0	1
Flood	=1 if experinaced flood during last four years	979	0.13	0.34	0	1	938	0.13	0.33	0	1
Hail storms/erosion	=1 if experinaced hail storms/erosion during last four years	979	0.12	0.32	0	1	974	0.16	0.36	0	1
Epidemic livestock and pests and crop disease	=1 if experinaced epidemic during last four years	979	0.27	0.44	0	1	974	0.37	0.48	0	1
Drought_comm	Cummunity level drought shock	979	0.25	0.43	0	1	974	0.14	0.34	0	1
Flood_comm	Community level flood shock	979	0.15	0.36	0	1	974	0.23	0.42	0	1
Hail/storm_comm	Community level hail storms	979	0.15	0.35	0	1	974	0.19	0.39	0	1
Crop_comm	Community level epidemics	979	0.14	0.34	0	1	974	0.23	0.42	0	1
Help_drought	=1 if help received after drought and zero otherwise	979	0.04	0.18	0	1	938	0.03	0.17	0	1
Help_flood	=1 id help recieved after flood and zero otherwise	979	0.01	0.07	0	1	938	0.014	0.12	0	1
Help_stroms	=1 if help received after hail storms and zero otherwise	979	0.01	0.09	0	1	974	0.01	0.09	0	1
Help_crop	=1 if help received after epidemics and zero otherwise	979	0.03	0.17	0	1	974	0.02	0.14	0	1
Help_sh_comm	=1 if help received after any community level shock and zero otherwise	979	0.69	0.46	0	1	974	0.58	0.49	0	1
Site type	=1 if urban and zero if rural	979	0.59	0.49	0	1	938	0.59	0.49	0	1
Lnpop	Log of population of community	957	9.09	0.61	7.94	10.75	884	8.97	0.72	7.73	11.15

Our first measure for learning is a categorical variable for whether or not each child is enrolled in school. While time in schooling as reflected in enrollment is probably an indication of learning, it is not learning per se. Also in our sample and in much of Sub-Saharan Africa there is not much variance in enrollments for the young adolescents whom we consider. The enrollment rate was very high (97%) in 2006 and 90% in 2009.

Our second and third measures for learning are cognitive skills as measured with scores achieved in Peabody picture vocabulary test¹ and mathematics tests. These are direct measures of some important dimensions of learning, which may be more affected by climatic shocks than school enrollment. For example, in response to such shocks children may continue to be enrolled in school but switch their time use towards more work, which may disrupt learning (Boyden et al., 1998: 249). Children may be too tired to concentrate in class or do homework. School may be scheduled at times of day difficult for working children and children may be punished for arriving at school late (Nieuwenhuys, 1994: 70). If children miss classes or days because of work, they may fall behind and become discouraged (Boyden et al., 1998: 256). If they start school late because they have been working, they may be embarrassed and frustrated by being older than their classmates (Boyden et al., 1998: 256). Thus we use two other measures, PPVT and math score, for child learning outcomes. Children of age 11–12 years in round 2 and age 14–15 years in round 3 took the PPVT in the language in which they felt most comfortable so they can perform to their maximum capability. Mathematics tests in both rounds consisted of two sections (with the first section including addition, subtraction, division and multiplication and the second section including data interpretation, problem solving and knowledge of geometry). The averages for the PPVT and mathematics tests are almost identical in both rounds, but the variations in both tests were higher in 2006.

Health and nutritional status are represented by anthropometric measures of body mass index (BMI) and height for age (HFA). Temperature and precipitation may affect the prevalence of vector borne diseases, water borne and water washed diseases, as well as determine heat or cold stress exposure (Confalonieri et al., 2007), all of which may have short-term or longer-term implications for health and nutritional status. BMI captures primarily short-term effects and height is usually assumed to represent longer-term effects. Under the assumption that healthy children follow similar growth patterns across different populations, children’s anthropometric measurements are standardized according to the International Reference Population defined by the U.S. National Center for Health Statistics (NCHS), with the Centers for Disease Control (CDC) and the World Health Organization (WHO). Therefore, anthropometric measurements are expressed as z-scores; that is, a child’s measurements are compared to those of child of the same age and gender in a healthy reference population that has a z-score with a mean of zero and a standard deviation of one. The Ethiopian children in our sample for the most part have a low distribution of anthropometric values, with means and most of the distributions substantially below the international reference standards. In our data almost 30% of the children were stunted (less than –2 HFA z-score) and almost 50% of the children in this cohort are underweight (less than –2 BMI z-score). This implies a generally poor nutritional status, both in the short-term (BMI) and longer-term (HFA).

To measure the climatic shocks we use four categorical variables: whether respondents have experienced during the last four years, in order of their prevalence as reported by the individual respondents, (1) drought (31% in 2006, 37% in 2009); (2) livestock epidemics and crop pests and diseases (27%, 37%) that may be affected by climatic variations among other factors; (3) hailstorms and land erosion (12%, 16%);² and (4) floods (13%, 13%).³ The respondents reported the first two at least twice as frequently as they reported the last two. In agriculture of the types in these villages, with considerable heterogeneity at the very local level with regard to topography and water conditions, individual shocks do not necessarily reflect community-level shocks. In this type of agriculture, these problems may affect one plot and not the next or they may affect many plots and thus be community-level. For this reason we also estimate the impact of these climatic shocks reported by community informants in addition to those reported at the individual level. The community informant reports, however, are more equal among these four shocks than are the individual reports: (1) drought (25%, 14%); (2) livestock epidemics and crop pests and diseases (14%, 23%); (3) hailstorms and land erosion (15%, 19%); and (4) floods (15%, 23%). This suggests that individual respondents generally perceive more difficulties in coping relatively with the first two of these than do community informants. To explore whether foreign interventions, government aid, and cash received recently from any source help buffer such climatic shocks, we include household responses about receiving help from these sources. The individual reports on such help in response to the various shocks are few, ranging from 1% to 4%.

In addition to climatic shocks and sources of resources for buffering the shock, we include as controls household characteristics (household wealth, parental schooling, household size), child characteristics (gender, a quadratic in age to capture life cycle patterns) and community characteristics (in population, whether rural).

Results and discussion

We estimate the five sets of equations, one each for the three indicators of learning and for the two indicators of health and nutritional status. In each set of equations, we estimate eight equations: one each of for each of the four individual-respondent-reported shocks and one each for each of the four community-informant-reported shocks. In each equation, as noted above, we control for household, child and community characteristics. In each equation we control for individual random effects because we are including observations for each individual for both round 2 and round 3.

Child learning outcomes

School enrollment

We find two significant effects in probit estimates of community-level shocks on school enrollment: negative for drought shocks (community-level) and positive for hailstorm/erosion (see Table 2). The opposite signs are not necessarily puzzling, but instead simply may reflect the dominance of income effects of droughts and of price effects for hailstorm/erosion.⁴ That is, for the latter case there may be a much greater drop in the value of child labor than in the former case. Were there data on child wage rates, it would be possible to see if they are consistent with such a pattern – but unfortunately, we do not have such data. Drought in Kenya caused many children to be out of school because during this time they helped their families and fetched ever-scarcer water from further away (Serna, 2011). Other studies of similar-aged children in other developing country contexts have also reported negative effects of drought on school enrollment (e.g. Alderman et al., 2006; Glewwe et al., 2000; Hoddinott and Kinsey, 2001), but we are not aware of other studies on the impacts of hailstorms/erosion.

Table 2.

Different climatic shocks and child enrollment (random effects).

	Drought	Drought_com	Flood	Flood_com	Hailstorms	Hailstorm_com	Crops/livestock	Crop/livestock_com
Age (months)	0.001	0.01	−0.004	−0.008	−0.012	−0.03	−0.006	−0.03
	[0.135]	[0.135]	[0.135]	[0.135]	[0.135]	[0.135]	[0.135]	[0.136]
Age sq.	−0.0001	−0.0001	−0.0001	−0.0001	−0.00003	0.000009	−0.00006	0.00001
	[0.0004]	[0.0004]	[0.0004]	[0.0004]	[0.0004]	[0.0004]	[0.0004]	[0.0004]
Sex	0.26*	0.28*	0.26*	0.26*	0.26*	0.26*	0.26*	0.26*
	[0.108]	[0.108]	[0.107]	[0.107]	[0.107]	[0.108]	[0.108]	[0.107]
Wi	1.56***	1.73***	1.68***	1.73***	1.71***	1.69***	1.66***	1.75***
	[0.468]	[0.469]	[0.462]	[0.476]	[0.461]	[0.467]	[0.459]	[0.467]
HH size	−0.031	−0.036	−0.036	−0.038	−0.036	−0.037	−0.034	−0.037
	[0.0273]	[0.0274]	[0.027]	[0.027]	[0.027]	[0.027]	[0.027]	[0.027]
Mother’s education (primary)	0.25	0.26	0.24	0.25	0.25	0.23	0.25	0.26
	[0.17]	[0.17]	[0.17]	[0.17]	[0.17]	[0.17]	[0.17]	[0.17]
Mother’s education (secondary)	−0.09	−0.09	−0.091	−0.0922	−0.0931	−0.145	−0.089	−0.068
	[0.129]	[0.129]	[0.129]	[0.129]	[0.129]	[0.132]	[0.129]	[0.131]
Father’s education (primary)	0.05	0.03	0.059	0.06	0.07	0.05	0.06	0.04
	[0.171]	[0.172]	[0.171]	[0.171]	[0.172]	[0.172]	[0.171]	[0.172]
Father’s education (secondary)	−0.08	−0.1	−0.06	−0.07	−0.07	−0.09	−0.07	−0.09
	[0.139]	[0.140]	[0.140]	[0.140]	[0.139]	[0.140]	[0.139]	[0.140]
Climatic shock^a	−0.09	−0.22*	0.09	−0.03	0.23	0.35*	0.003	0.14
	[0.120]	[0.128]	[0.154]	[0.145]	[0.162]	[0.157]	[0.116]	[0.129]
Help received^b	−0.19	0.09	0.35	0.07	−0.35	0.095	−0.052	0.06
	[0.237]	[0.110]	[0.535]	[0.116]	[0.454]	[0.110]	[0.300]	[0.111]
Typesite – rural	−0.24	−0.22	−0.29	−0.27	−0.31*	−0.3*	−0.28	−0.29
	[0.193]	[0.193]	[0.190]	[0.190]	[0.191]	[0.191]	[0.195]	[0.189]
Ln(pop)	0.14	0.11	0.13	0.118	0.107	0.09	0.13	0.14
	[0.110]	[0.110]	[0.109]	[0.110]	[0.109]	[0.110]	[0.109]	[0.112]
_cons	1.32	1.46	2.55	2.92	3.43	4.69	2.69	4.39
	[10.98]	[10.96]	[10.95]	[10.98]	[10.98]	[11.02]	[10.95]	[11.05]
N	1813	1813	1813	1813	1813	1813	1813	1813

Note: Standard errors in brackets

p<0.05, **p<0.01, *** p<0.001

Climatic shock shows the impact type of the climatic event mentioned in the first horizontal row of the table.

See table 1 for definition of variables.

We find no significant effects of individual-reported shocks or of reported help received on enrollment in school. A plausible explanation for the latter result is than any help received after droughts is not sufficient or is not well-targeted (Demombynes and Kiringai, 2011; Webb and Reardon, 1992). However, results also show that stock of physical and human capital has a positive impact on child schooling. Yamauchi et al. (2009) report that in Malawi and Ethiopia intellectual human and physical capital before disaster helps the household to maintain schooling investment in children after natural disaster. Shah and Steinberg (2013) argue that school attendance during drought years depends on outside options, particularly the wage rate in the labor market in India.

PPVT test scores

Droughts and hailstorms/erosion have significant negative effects on PPVT scores, whether the shocks are reported by individuals or by community informants (see Table 3). The negative sign for community-informant-reported hailstorms/erosion is in contrast to the significant positive effect of this variable on school enrollment. Taken together, these two coefficient estimates suggest that due to the reduction of work options children are more likely to enroll in school, but nevertheless they learn less, perhaps because of other pressures on their time (e.g. home repairs) even if labor markets are depressed. The coefficient estimate for individual-reported floods is also negative, similar to those for droughts and hailstorms/erosion. Sprung and Harris (2010) find that there is a shift towards negative content following hurricane exposure compared with non-hurricane exposed children, and knowledge about thinking is linked to the reporting of such intrusive thoughts. However, the coefficient estimates for community-respondent-reported flood and livestock/crop disease shocks are both positive, suggesting that the price effects of lessened labor market alternatives dominate so that learning increases due to more time studying. Shah and Steinberg (2013) show that rainfall shock in early life has a positive impact, and results in children scoring .03 points higher on math and reading tests today in developing countries. Help received has a significant positive coefficient estimate for individual-reported crop/livestock pest/disease shocks, consistent in this case with such help relieving the negative income impact of the shocks. Conversely, help received has significant negative coefficient estimates for individual-reported drought and hailstorm/erosion shocks. While it is possible that such help enabled the afflicted households to reconstruct better their productivity activities and this put negative time pressure on the adolescent children, this seems to be a somewhat convoluted explanation. Rossel (2008) reports that the negative coefficient of the interacted variables (climatic shock and help received) suggests that households which received aid appear to be worse off than households which didn’t.

Table 3.

Different climatic hocks: shock and PPVT (random effects).

	Drought	Drought_com	Flood	Flood_com	Hailstorms	Hailstorm_com	Crops/livestock	Crop/livestock_com
Age (months)	5.67*	2.35	6.04*	7.14**	6.10**	7.67**	5.52*	6.23*
	[2.617]	[2.584]	[2.607]	[2.585]	[2.619]	[2.626]	[2.612]	[2.611]
Age sq.	−0.02*	−0.01	−0.02*	−0.02**	−0.02**	−0.02**	−0.0169*	−0.0190*
	[0.008]	[0.007]	[0.007]	[0.007]	[0.008]	[0.008]	[0.008]	[0.008]
Sex	−3.66*	−4.08*	−3.9*	−3.9*	−4.3*	−3.6	−3.9*	−4.08*
	[2.185]	[2.155]	[2.199]	[2.210]	[2.191]	[2.212]	[2.189]	[2.188]
Wi	42.5***	49.5***	43.4***	43.3***	44.7***	46.6***	43.8***	46.3***
	[8.664]	[8.475]	[8.651]	[8.677]	[8.613]	[8.672]	[8.611]	[8.625]
HH size	−0.58	−0.41	−0.52	−0.78	−0.64	−0.86	−0.78	−0.87
	[0.536]	[0.533]	[0.541]	[0.544]	[0.535]	[0.544]	[0.535]	[0.541]
Mother’s education (primary)	4.27	6.73**	3.31	2.78	3.43	2.34	3.22	2.82
	[3.113]	[3.095]	[3.111]	[3.126]	[3.103]	[3.128]	[3.093]	[3.101]
Mother’s education (secondary)	3.665	5.0*	3.094	2.872	3.099	2.486	2.381	2.470
	[2.842]	[2.800]	[2.835]	[2.847]	[2.834]	[2.844]	[2.815]	[2.826]
Father’s education (primary)	4.613	3.483	5.084	6.16*	4.773	5.328	5.585	4.835
	[3.432]	[3.376]	[3.447]	[3.448]	[3.448]	[3.447]	[3.430]	[3.433]
Father’s education (secondary)	4.01	3.37	4.09	5.13*	4.07	4.21	4.73	3.71
	[3.011]	[2.953]	[3.022]	[3.023]	[3.028]	[3.024]	[3.016]	[3.021]
Climatic shock^a	−6.82*	−35.4***	−8.7**	10.7***	−7.73**	−8.83***	0.7	14.01***
	[3.25]	[4.84]	[3.55]	[2.63]	[3.58]	[2.59]	[3.03]	[4.09]
Help received^b	−30.1*	−3.97	−27.47	0.89	−53.76*	2.03	−46.36***	3.15
	[11.76]	[2.456]	[17.87]	[2.416]	[30.60]	[2.443]	[12.63]	[2.505]
Typesite – rural	−18.2***	−13.8***	−19.5***	−19.8***	−19.0***	−17.5***	−21.3***	−21.0***
	[3.58]	[3.52]	[3.48]	[3.47]	[3.53]	[3.62]	[3.55]	[3.43]
Ln (pop)	8.2***	8.99***	8.6***	6.54**	8.48***	9.08***	8.41***	9.47***
	[2.02]	[2.0]	[2.03]	[2.09]	[2.02]	[2.07]	[2.02]	[2.07]
_cons	−240.1	30.1	−272.7	−343.4*	−276.0	−410.5*	−227.1	−298.5
	[212.1]	[209.5]	[211.2]	[209.3]	[212.3]	[213.2]	[211.8]	[211.6]
N	833	833	833	833	833	833	833	833

Note: Standard errors in brackets

p<0.05, **p<0.01, *** p<0.001

Climatic shock shows the impact type of the climatic event mentioned in the first horizontal row of the table.

See table 1 for definition of variables.

Mathematics test scores

The only significant shock coefficient shown in mathematics test scores is for community-informant-reported hailstorms/erosion, which is negative as is the case for PPVT discussed above (but, as also discussed above, is the opposite from the sign for school enrollment) (see Table 4). For the individual-reported crop/livestock pest/disease shock, help received has a significant negative coefficient estimate – parallel to two cases for PPVT mentioned above for which any explanation that we can come up with seems convoluted.

Table 4.

Different climatic shocks: shock and math score (random effects).

	Drought	Drought_com	Flood	Flood_com	Hailstorms	Hailstorm_com	Crops/livestock	Crop/livestock_com
Age (months)	0.63	0.46	0.53	0.64	0.67	0.69	0.18	0.28
	[2.067]	[2.069]	[2.075]	[2.061]	[2.073]	[2.065]	[2.067]	[2.072]
Age sq.	−0.002	−0.002	−0.002	−0.002	−0.002	−0.002	−0.001	−0.001
	[0.006]	[0.006]	[0.006]	[0.006]	[0.006]	[0.006]	[0.006]	[0.006]
Sex	−3.9*	−4.03*	−4.004*	−4.03*	−4.06*	−3.84*	−4.34*	−4.04*
	[2.04]	[2.036]	[2.034]	[2.045]	[2.029]	[2.037]	[2.035]	[2.032]
Wi	32.5***	34.1***	34.01***	32.6***	33.6***	34.4***	33.9***	34.4***
	[8.053]	[7.998]	[7.977]	[8.074]	[7.972]	[7.990]	[7.963]	[7.991]
HHsize	−0.22	−0.29	−0.28	−0.32	−0.3	−0.35	−0.25	−0.31
	[0.488]	[0.491]	[0.487]	[0.491]	[0.485]	[0.490]	[0.486]	[0.489]
Mother’s education (primary)	4.65*	4.71*	4.67*	4.39	4.51	4.76*	4.32	4.99*
	[2.858]	[2.864]	[2.856]	[2.872]	[2.851]	[2.855]	[2.854]	[2.862]
Mother’s education (secondary)	6.37**	6.47**	6.51**	6.36*	6.36*	7.07**	6.19*	6.8**
	[2.490]	[2.491]	[2.492]	[2.497]	[2.487]	[2.503]	[2.486]	[2.497]
Father’s education (primary)	10.1***	10.3***	10.3***	10.4***	10.2***	10.3***	10.1***	10.07***
	[3.015]	[3.017]	[3.013]	[3.019]	[3.010]	[3.010]	[3.007]	[3.015]
Father’s education (secondary)	6.85**	6.8**	6.87**	7.2**	6.85**	6.82**	6.82**	6.46*
	[2.569]	[2.578]	[2.573]	[2.577]	[2.571]	[2.566]	[2.567]	[2.585]
Climatic shock^a	−2.68	−0.85	−1.83	3.45	−4.62	−5.28**	−0.97	3.72
	[2.346]	[2.485]	[3.066]	[2.444]	[3.208]	[2.343]	[2.266]	[2.494]
Help received^b	−6.56	0.23	3.65	0.49	18.72	0.14	−14.66*	0.45
	[5.604]	[1.918]	[15.79]	[1.923]	[14.55]	[1.914]	[6.593]	[1.921]
Typesite – rural	0.201	−0.931	−0.896	−1.105	−0.541	−0.332	−0.497	−1.616
	[3.490]	[3.424]	[3.389]	[3.369]	[3.395]	[3.377]	[3.436]	[3.370]
Ln (pop)	3.717*	3.405*	3.46*	3.198	3.67*	3.8*	3.42*	3.45*
	[1.957]	[1.958]	[1.942]	[1.965]	[1.946]	[1.963]	[1.937]	[1.956]
_cons	210.9	226.2	220.6	214.9	206.9	203.4	249.4	239.4
	[167.5]	[167.4]	[168.0]	[166.7]	[167.9]	[167.3]	[167.3]	[167.7]
N	1522	1522	1522	1522	1522	1522	1522	1522

Note: Standard errors in brackets

p<0.05, **p<0.01, *** p<0.001

Climatic shock shows the impact type of the climatic event mentioned in the first horizontal row of the table.

See table 1 for definition of variables.

Our results confirm that physical assets (wealth index) and household human capital (parent’s education) play a positive and significant role in buffering the shock in all equations for learning outcomes which include school enrollment, PPVT test score and mathematics test score. Hoddinott (2004) reveals in the case of Zimbabwe that households with higher levels of asset holdings may choose to cope with a shock by selling some assets. The paper further elaborates that their decision to sell household assets does not necessarily carry with it a cost to the family’s future earnings or consumption. Nor does it necessarily preclude the household’s ability to recover these assets at a later time.

Child health and nutritional status

BMI z-scores

Individual-reported drought and hailstorms/erosion shocks are significantly negatively associated with the indicator of short-run health and nutritional status, as is the community-informant-reported crop/livestock pest/disease shock (see Table 5). However, the community-informant-reported hailstorm/erosion shock is significantly positively associated, which would seem to be the opposite of what would be expected if there were a pure income effect. Perhaps this reflects associated temperature cooling with this shock, as this may induce increased short-run consumption and less infectious disease risk so that there are short-term weight gains.

Table 5.

Different climatic shocks: shock and BMI z-Score (random effects).

	Drought	Drought_com	Flood	Flood_com	Hailstorms	Hailstorm_com	Crops/livestock	Crop/livestock_com
Age (months)	−0.06	−0.07*	−0.07	−0.07*	−0.07	−0.08*	−0.07*	−0.06
	[0.0439]	[0.0439]	[0.0439]	[0.0440]	[0.0439]	[0.0437]	[0.0439]	[0.0438]
Age sq.	0.0002	0.0002	0.0002	0.0002	0.0002	0.0002*	0.0002	0.0002
	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]
Sex	0.3***	0.3***	0.3***	0.3***	0.3***	0.3***	0.3***	0.3***
	[0.064]	[0.064]	[0.064]	[0.064]	[0.064]	[0.064]	[0.064]	[0.064]
Wi	0.5*	0.58**	0.59**	0.59**	0.58**	0.59**	0.6**	0.59**
	[0.22]	[0.22]	[0.22]	[0.22]	[0.22]	[0.22]	[0.22]	[0.22]
HH size	0.01	0.012	0.012	0.013	0.012	0.01	0.012	0.012
	[0.014]	[0.014]	[0.014]	[0.014]	[0.014]	[0.014]	[0.014]	[0.014]
Mother’s education (primary)	0.12*	0.12	0.13	0.13	0.13	0.12	0.12	0.12
	[0.083]	[0.083]	[0.083]	[0.083]	[0.083]	[0.083]	[0.083]	[0.083]
Mother’s education (secondary)	0.23**	0.24***	0.24***	0.24***	0.24***	0.22**	0.23**	0.23**
	[0.073]	[0.073]	[0.073]	[0.073]	[0.073]	[0.073]	[0.073]	[0.073]
Father’s education (primary)	−0.1	−0.1	−0.1	−0.1	−0.1	−0.1	−0.1	−0.1
	[0.08]	[0.08]	[0.08]	[0.08]	[0.08]	[0.08]	[0.08]	[0.08]
Father’s education (secondary)	−0.1	−0.1	−0.1	−0.1	−0.1	−0.1	−0.1	−0.1
	[0.068]	[0.068]	[0.068]	[0.068]	[0.068]	[0.067]	[0.068]	[0.068]
Climatic shock^a	−0.12*	0.07	−0.03	−0.03	−0.12*	0.16**	−0.03	−0.14*
	[0.055]	[0.062]	[0.068]	[0.064]	[0.073]	[0.054]	[0.054]	[0.056]
Help received^b	0.02	−0.03	−0.02	−0.03	0.1	−0.02	−0.2	−0.02
	[0.122]	[0.0432]	[0.215]	[0.0447]	[0.229]	[0.0426]	[0.141]	[0.0427]
Typesite – rural	−0.33***	-0.4***	−0.4***	−0.4***	−0.3***	−0.4***	−0.3***	−0.4***
	[0.096]	[0.096]	[0.096]	[0.096]	[0.096]	[0.096]	[0.097]	[0.096]
Ln (pop)	−0.2**	−0.2**	−0.2**	−0.16**	−0.15**	−0.18***	−0.16**	−0.18***
	[0.052]	[0.053]	[0.052]	[0.052]	[0.052]	[0.052]	[0.052]	[0.052]
_cons	4.82	5.44	5.26	5.49	5.19	6.4*	5.63	4.79
	[3.583]	[3.580]	[3.582]	[3.589]	[3.584]	[3.576]	[3.586]	[3.574]
N	1836	1836	1836	1836	1836	1836	1836	1836

Note: Standard errors in brackets

p<0.05, **p<0.01, *** p<0.001

Climatic shock shows the impact type of the climatic event mentioned in the first horizontal row of the table.

This variable shows the help received after a particular natural disaster.

Table 6.

Different climatic shocks: shock and HFA z-score (random effects).

	Drought	Drought_com	Flood	Flood_com	Hailstorms	Hailstorm_com	Crops/livestock	Crop/livestock_com
Age (months)	−0.1***	−0.1***	−0.1***	−0.12**	−0.1	−0.13**	−0.14***	−0.14***
	[0.04]	[0.04]	[0.04]	[0.04]	[0.04]	[0.04]	[0.04]	[0.04]
Age sq.	0.001**	0.001**	0.001***	0.0003**	0.0002	0.0004**	0.0004***	0.0004***
	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]	[0.0001]
Sex	0.5***	0.5***	0.5***	0.5***	0.3***	0.5***	0.5***	0.5***
	[0.07]	[0.07]	[0.07]	[0.07]	[0.063]	[0.07]	[0.07]	[0.07]
Wi	0.809***	0.809***	0.812***	0.729**	0.579**	0.782***	0.810***	0.792***
	[0.225]	[0.225]	[0.225]	[0.225]	[0.220]	[0.225]	[0.225]	[0.225]
HH size	−0.001	−0.001	−0.002	−0.003	0.012	−0.001	−0.002	0.00001
	[0.015]	[0.015]	[0.015]	[0.015]	[0.014]	[0.014]	[0.014]	[0.015]
Mother’s education (primary)	−0.02	−0.02	−0.03	−0.05	0.13	−0.03	−0.03	−0.027
	[0.091]	[0.091]	[0.091]	[0.091]	[0.08]	[0.091]	[0.091]	[0.091]
Mother’s education (secondary)	0.06	0.06	0.06	0.06	0.24***	0.08	0.06	0.07
	[0.076]	[0.076]	[0.077]	[0.076]	[0.073]	[0.076]	[0.076]	[0.077]
Father’s education (primary)	0.13	0.13	0.14*	0.15*	−0.11	0.15*	0.15*	0.14
	[0.086]	[0.086]	[0.086]	[0.086]	[0.083]	[0.086]	[0.086]	[0.086]
Father’s education (secondary)	0.13*	0.13*	0.143*	0.15**	−0.109	0.14**	0.145*	0.145*
	[0.0681]	[0.0681]	[0.0683]	[0.0679]	[0.0682]	[0.0680]	[0.0682]	[0.0681]
Climatic shock^a	−0.2**	−0.2**	−0.001	0.22***	−0.12*	−0.11*	0.01	0.04
	[0.061]	[0.061]	[0.065]	[0.063]	[0.073]	[0.051]	[0.051]	[0.054]
Help received^b	−0.02	−0.02	−0.04	0.01	0.15	−0.03	−0.03	−0.03
	[0.041]	[0.041]	[0.203]	[0.042]	[0.229]	[0.04]	[0.134]	[0.04]
Typesite – rural	−0.15	−0.15	−0.15	−0.18*	−0.35***	−0.15	−0.16	−0.17
	[0.102]	[0.102]	[0.102]	[0.102]	[0.0963]	[0.103]	[0.103]	[0.103]
Ln (pop)	0.052	0.052	0.085	0.056	−0.15**	0.09*	0.085	0.082
	[0.053]	[0.053]	[0.052]	[0.053]	[0.052]	[0.053]	[0.052]	[0.053]
_cons	7.9*	7.9*	8.0*	7.12*	5.19	7.34*	8.05*	8.2*
	[3.278]	[3.278]	[3.280]	[3.284]	[3.584]	[3.293]	[3.282]	[3.286]
N	1836	1836	1836	1836	1836	1836	1836	1836

Note: Standard errors in brackets

p<0.05, **p<0.01, *** p<0.001

Climatic shock shows the impact type of the climatic event mentioned in the first horizontal row of the table.

See table 1 for definition of variables.

The log of community population is negative and significant in all BMI z-score equations. The increasing population in Ethiopia is also another hurdle in child health due to limited resources. The population of sub-Saharan Africa is increasing, and the escalating urban population in African cities and towns poses a real challenge to managing disaster risk (Pelling and Wisner, 2009).

HFA z-scores

Individual-reported and community-informant-reported drought and hailstorm/erosion shocks are all significantly negatively associated with this measure of long-term nutritional status, thus implying long-term negative effects on health and nutritional status resulting from climatic shocks in the last four years during late childhood and early adolescence. Such results are consistent with some previous literature. For example, Yamano et al. (2005) find, also for Ethiopia, that a 10% increase in the proportion of damaged crop areas corresponded to a reduction in child growth of 0.12 cm over a six-month period. Rossel (2008) reports that children in Peru affected by a climatic shock have approximately a 0.16 lower ZHFZ than other children at age five. However, the association with the community-informant-reported flood shock is positively significant, which is somewhat puzzling.

Conclusion

Poor households, particularly poor households in rural areas, are widely perceived as being quite vulnerable to climatic shocks. In this paper we investigated the associations between such shocks and adolescent human capital indicators in the very poor context of Ethiopia. We find evidence of significant and in many cases substantial effects of individual and community-informant-reported shocks on some basic indicators of the learning, health and nutritional status of adolescents. Most of these effects are negative, but some cases of positive effects are also observable. This reflects at least in a subset of cases the dominance of price over income effects. Our results support the argument developed by Ferreira and Schady (2008) that the economics of idiosyncratic and aggregate shocks is different. Tables 7 and 8 show the direction of shock-effects on child outcomes. The results show that all idiosyncratic shocks have a negative impact but few aggregate or community shocks have positive impact due to the dominance of the substitution effect.

Table 7.

Summary table of shocks with negative impacts on child outcomes.

	Child outcomes
Climatic shock	Enrollment	PPVT test	Math score	BMI z-score	HFA z-score
Drought		Negative		Negative
Drought_com	Negative	Negative			Negative
Flood		Negative
Hailstorms		Negative		Negative	Negative
Hailstorms_com		Negative	Negative		Negative
Crops/livestock_com				Negative

PPVT: Peabody picture vocabulary test; BMI: body mass index; HFA: height for age

Table 8.

Summary Table of Shocks with Positive Impact on Child Outcomes.

	Child outcomes
Climatic shock	Enrollment	PPVT test	Math score	BMI z-score	HFA z-score
Flood_com		Positive			Positive
Hailstorms_com	Positive			Positive
Crops/livestock_com		Positive

PPVT: Peabody picture vocabulary test; BMI: body mass index; HFA: height for age

Girls are better off in both health outcomes (boys are more stunted in Africa (Araujo et al., 2012; Choudhury and Bhuiya, 1993; Marcoux, 2002; Svedberg, 1990; Wamani et al., 2007)) and worse off in PPVT and mathematics scores. Household human capital (parent’s education) and household physical capital (wealth index) have positive impacts on child investment (the same findings are reported by Araujo et al. (2012) for Burkina Faso).

Rural areas in developing countries are especially vulnerable and lack the necessary coping and adaptive mechanism to deal with extreme temperature shocks (IPCC, 2001). Our estimates confirm that being in a rural area has a significant and negative effect on all kinds of learning and health outcomes. Help received by the shock-affected household is irrelevant (the same results are reported by Rossel (2008) for Peruvian data) except for PPVT and mathematics test scores, where “help received” appears as negative and significant (Sann el al. (2012) found the same result for Cambodia). Reasons for this result include: aid is not sufficient to support the learning process and usually households use the aid for nutritional purposes or for investment in land in order to get an immediate return. Moreover, in this situation children shift their efforts and attention from schooling to agricultural work.

The findings carry an important message for policy makers and international donors, particularly during climatic shocks. Our summary statistics and regression analysis shows that boys are more stunted in Ethiopia, and girls, though better off in health status, lag behind boys on cognitive skills. Thus, health and schooling policies can be targeted on the basis of gender for more effective outcomes. The main findings of the study indicate that idiosyncratic shocks have a significant negative effect on child learning, and health outcomes can be addressed by parents’ access to the credit market and medical insurance. However, the positive effect of few community or aggregate shocks adversely affects the quality of education because of the greater number of students in schools. Thus, in such circumstances the foreign or government intervention programs can be targeted toward schools rather than individual households. Another very important finding is the insignificance of the “help received” variables, which include foreign or governmental aid or help during or after the climatic crisis. It has been observed in many developing countries that whenever aid agencies or cash transfer programs are about to reach a community, children are intentionally taken out of schools by the parents to ensure their eligibility for cash transfer programs (see for example De Janvry et al., 2004). A very careful policy is required in these programs in order to discourage such parental behavior.

Further, the aid programs must be shock-specific and structured according to the nature of climatic shock. Different hazards cause different impacts. Therefore, the association of household responses to hazards must be better understood in order to target policy and resource allocation. According to the Climate Change Report (2013) it is very likely that the entire African continent will continue to warm during the 21st century. Further, the report also predicts the reduction in the long rains over Kenya and Ethiopia in response to warmer Indian Ocean sea surface temperatures. Therefore, precautionary measures should be undertaken in poor communities, which are more vulnerable during climatic shocks.

The paper also opens many research avenues; for example, the climatic shocks may explain household behavior like migration or relocation, which may lead to different child outcomes. The length of shock and its impact in the short-term and long-term is yet another area to be explored. Controlling for lags and leads will be helpful for more dynamic analysis. Finally, coping strategies, both at the micro and macro level, to buffer the climatic shocks, is another important area of research which may need further investigation.

Footnotes

Notes

Author biographies

Asma Hyder Balochis teaching at National University of Sciences and Technology, Islamabad. She is recipient of Fulbright J. William Award for International Understanding and did her Post Doctorate from University of Pennsylvania, USA. She is also a research affiliate at Institute for Research on Labor and Employment, University of California Los Angeles. Her research interest focuses on labor economics, wage differentials, schooling, economic shocks and gender.

Jere R Behrman is the WR Kenan, Jr. Professor of Economics and Sociology and Research Associate of the Population Studies Center at the University of Pennsylvania. His research is in empirical micro economics, economic development, labor economics, human resources (education, training, health, nutrition), economic demography, household behaviors, life-cycle and intergenerational relations and policy evaluation. He has published over 350 professional articles (primarily in leading general and field economic journals, also in leading demographic, sociology, nutritional and biomedical journals) and 33 books. He has been a research consultant with a number of international organizations, involved in professional research or lecturing in over 40 countries, a principal investigator on over 70 research projects and received a number of honors for his research.

References

Alderman

Hoddinott

Kinsey

(2006) Long-term consequences of early childhood malnutrition. Oxford Economic Papers 58(3): 450–474.

Araujo Bonjean

Brunelin

Simonet

(2012) Impact of climate related shocks on child’s health in Burkina Faso, CERDI, Etudes et Documents, E 2012.32.

Asmare

Berhane

(2013) Effect of climatic shock on cognitive development of children in Ethiopia. Eleventh International Conference on Ethiopia Economy, Addis Ababa, Ethiopia, July 18–20, 2013.

Baez

Santos

(2007) Children’s vulnerability to shocks: Hurricane Mitch in Nicaragua as a natural experiment. Paper presented on seminar at Maxwell School of Citizenship and Public Affairs, Syracuse University and Kennedy School of Government Harvard University.

Bau

(1999) Providing culturally competent services to visually impaired persons. Journal of Visual Impairment & Blindness 93: 291–297.

Beegle

Weerdt

De J

Dercon

(2006) “Kagera Health and Development Survey 2004 Basic Information Document,” World Bank.

Behrman

J R.

David

Richard

, 2008. “Improving quality versus increasing the quantity of schooling: Estimates of rates of return from rural Pakistan,” Journal of Development Economics, Elsevier, 85(1–2): 94–104.

Behrman

Murphy

Quisumbing

. (2009) Are returns to mothers’ human capital realized in the next generation? The impact of mothers’ intellectual human capital and long-run nutritional status on children’s human capital in Guatemala. Discussion Paper 850, Washington, D.C.: International Food Policy Research Institute.

Boyden

Ling

Myers

(1998) What Works for Working Children. Stockholm: RaddaBarnen.

10.

Burgess

Deschenes

Donaldson

. (2011) Weather and death in India. Unpublished working paper.

11.

Butt

McCarl

Angerer

(2005) The economic and food security implications of climate change in Mali. Climatic Change 68(3): 355–378.

12.

Cengage Learning (2005) Language Development, 69.

13.

Center for International Disaster Information (2002) International Disaster Situation Report. Available at: http://www.cidi.org/disaster. Retrieved on Jan 30, 2012.

14.

Choudhury

Bhuiya

(1993) Effects of Biosocial Variables on Changes in Nutritional Status of Rural Bangladeshi Children, Pre- and Post-Monsoon Flooding. Journal of Biosocial Science 25(3): 351–357.

15.

Climate Change Report (2013) Climate Change 2013: The Physical Science Basis. Intergovernmental Panel on Climate Change. Technical Support Unit, Working Group 1.

16.

Confalonieri

Menne

Akhtar

. (2007) Human health. In Climate Change 2007: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press.

17.

Crookston

Dearden

Alder

. (2010a) Impact of early and concurrent stunting on cognition. Maternal & Child Nutrition 7(4): 397–409.

18.

Crookston

Penny

Alder

. (2010b). Children who recover from early stunting and children who are not stunted demonstrate similar levels of cognition. The Journal of Nutrition, 1 Vol. 40(11): 1996–2001.

19.

De la Fuente

(2010) Remittances and vulnerability to poverty in rural Mexico. World Development 38: 828–839.

20.

De Janvry

Finan

Sadoulet

(2004) Can Conditional Cash Transfers Serve as Safety Nets to Keep Children at School and Out of the Labor Market? Department of Agricultural & Resource Economics, UCB. UC Berkeley: Department of Agricultural and Resource Economics. Available at: http://escholarship.org/uc/item/5fp0g5p2, Retrieved on 14 Jan, 2012.

21.

De Waal

Taffesse

Carruth

(2006) Child survival during the 2002–03 drought in Ethiopia. Global Public Health 1(2): 125–132.

22.

Desai

Chase-Lansdale

Michael

(1989) Mother or Market? Effects of Maternal Employment on the Intellectual Ability of 4-Year Old Children. Demography 26: 545–561.

23.

Dehejia

Lleras-Muney

(2004) Booms, busts, and babies’ health. Quarterly Journal of Economics 119: 1091–1130.

24.

Demombynes

Kiringai

(2011) The Drought and Food Crisis in the Horn of Africa: Impacts and Proposed Policy Responses for Kenya. Washington, DC: World Bank Poverty Reduction and Economic Management Network.

25.

Duc

(2009) The Effects of Early Age Stunting on Cognitive Achievement Among Children in Veitnam. Young Lives. Working Paper No. 45.

26.

Duc

Behrman

(2013) Early Childhood Stunting, Post-Infant Recovery and Educational Outcomes in Late Childhood: Evidence from Vietnam. Hanoi, Vietnam: Centre for Analysis and Forecasting, Vietnam Academy of Social Sciences.

27.

Duryea

Arends-Kuenning

(2003) “School Attendance, Child Labor, and Local Labor Markets in Urban Brazil.” World Development 31(7): 1165–1178.

28.

Escobal

Saavedra

Suárez

Huttly

Penny

Lanata

Villar

(2005) The Interaction of Public Assets, Private Assets and Community Characteristics and its Effect on Early Childhood Height-for-Age in Peru, Young Lives Working Paper 14, Lima: Young Lives Project.

29.

Ferreira

Francisco H.G.

Schady

Norbert

(2009) “Social Consequences of the Global Financial Crisis inLatin America : Some Preliminary, and Surprisingly Optimistic, Conjectures,” World Bank Other Operational Studies 10957, The World Bank.

30.

Flug

Spilimbergo

Wachtenheim

(1998)

Investment in education: do economic volatility and credit constraints matter?

Journal of Development Economics 55: 465–481.

31.

Hoddinott

Emmanuel

(2004) “The Impact of PROGRESA on Food Consumption,” Economic Development and Cultural Change, University of Chicago Press, 53(1): 37–61.

32.

Fallon

P R

Lucas

R EB

(2000) “The Impact of Financial Crises on Labor Markets, Household Incomes and Poverty: A Review of Evidence,” Boston University - Institute for Economic Development, Boston University, Institute for Economic Development 103, Boston University, Institute for Economic Development.

33.

Ferreira

FHG

Schady

(2008) Aggregate economic shocks, child schooling and child health. World Bank, Policy Research Working Paper No. 4701.

34.

Gajigo

Schwab

(2012) The rhythm of the rains: Seasonal effects on child health in The Gambia. 2012 Conference, August 18–24, 2012, Foz do Iguacu, Brazil 125788, International Association of Agricultural Economists.

35.

Garbero

Muttarak

(2012) Effects of the 2010 droughts and floods on community welfare in rural Thailand: Differential effects of village educational attainment. Available at: http://paa2013.princ-eton.edu/papers/131356, Retrieved on Jan 10, 2013.

36.

Glewwe

Jacoby

King

(2000) Early childhood nutrition and academic achievement: A longitudinal analysis. Journal of Public Economics 81: 345–368.

37.

Gommes

Petrassi

(1996) Rainfall variability and drought in Sub-Saharan Africa since 1960. FAO Agrometeorology Series, Working Paper No. 9.

38.

Hoddinott

Kinsey

(2001) Child growth in the time of drought. Oxford Bulletin of Economics and Statistics 63: 409–436.

39.

Intergovernmental Panel on Climate Change (IPCC) (2001) Third Assessment Report: Climate Change 2001: Impacts, Adaptation, and Vulnerability. Cambridge, UK: Cambridge University Press.

40.

International Strategy for Disaster Reduction (2004) Living with Risk: A Global Review of Disaster Reduction Initiatives. Vol. 2.

41.

Jacoby

Skoufias

(1997) Risk, Financial Markets, and Human Capital in a Developing Country, Review of Economic Studies, LXIV 311–335. (1999), 50–61.

42.

Maccini

Yang

(2009) Under the weather: Health, schooling, and economic consequences of early-life rainfall. American Economic Review 99(3): 1006–1026.

43.

McCulloch

Joshi

(2002) Child development and family resources: Evidence from the second generation of the 1958 British birth cohort. Journal of Population Economics 15: 283–304

44.

Marcoux

(2002) Sex differentials in undernutrition: A look at survey evidence. Population and Development Review 28(2): 275–284.

45.

Mulugeta

(2009) Ethiopia: Erosion Degrades Farm Productivity. Ethiopian Review, OXFOM.

46.

Nelson

(2009) Climate Change: Impact on Agriculture and Costs of Adaptation. Washington, DC: International Food Policy Research Institute.

47.

National Environment Management Authority (1998) State of the environment report for Uganda 1997/1998. National Environment Management Authority, Kampala, Uganda.

48.

Nair

(2009) Neonatal Screening Program for G6PD Deficiency in India: Need and Feasibility. Indian Pediatrics 46(12): 1045–49.

49.

Nieuwenhuys

(1994) Children’s Lifeworlds: Gender, Welfare and Labour in the Developing World. London and New York: Routledge.

50.

Pelling

Wisner

(2009) Disaster risk reduction cases from urban Africa. London: Earthscan.

51.

Rossel Deustua

(2008) The impact of climatic shocks on child nutrition in Peru. Young Lives, Student Paper, Oxford University.

52.

Ruhm

(2003) Good times make you sick. Journal of Health Economics 24(4): 637–658.

53.

Ruhm

(2005) Healthy living in hard times. Journal of Health Economics 24: 341–363.

54.

Sann

Oum

Kan

. (2012) Opportunities and challenges for Asian economies in the new millennium. The 13th International Convention of the East Asian Economic Association, Grand Copthorne Waterfront Hotel Singapore, October 19–20, 2012.

55.

Sawada

Lokshin

(1999) ‘Household Schooling Decisions in Rural Pakistan.’ World Bank Policy Research Working Paper No. 2541 (November).

56.

Seballos

Tanner

Tarazona

Gallegos

(2011) Children in a Changing Climate Research Report. Brighton: IDS; Children and Disasters: Understanding Impact and Enabling Agency.

57.

Schlenker

Lobell

(2010) Robust negative impacts of climate change on African agriculture. Environmental Research Letters 5(1): 014010.

58.

Serna

(2011) Northern Eastern Kenya drought assessment report. Save the Children.

59.

Shah

Steinberg

(2013) Drought of opportunities: Contemporaneous and long term impacts of rainfall shocks on human capital. NBER Working Papers 19140, National Bureau of Economic Research, Inc.

60.

Shitarek

(2012) Ethiopia Country Report, UK. http://www.dfid.gov.uk/Documents/publications1/Econ-Ear-Rec-Res-Full-Report%20.pdf, Retrieved on Feb 10, 2012.

61.

Skoufias

(2007) Poverty alleviation and consumption insurance: Evidence from PROGRESA in Mexico. Journal of Socio-Economics 36: 630–649.

62.

Skoufias

Vinha

(2012) Climate variability and child height in rural Mexico. Economics and Human Biology 10(1): 54–73.

63.

Sprung

Harris

(2010) Intrusive thoughts and young children’s knowledge about thinking following a natural disaster. Journal of Child Psychology and Psychiatry 51(10): 1115–1124.

64.

State of the Environment Report for Uganda (1998) NEMA (National Environmental Management Authority), Kampala.

65.

Svedberg

(1990) Undernutrition in sub-Saharan Africa: Is there a gender bias? Journal of Development Studies 26(3): 469–486.

66.

United Nations Development Program (2011) Human Development Report 2011: Sustainability and Equity: A Better Future for All. New York: United Nations.

67.

UNICEF (2011) The Impact of the Food and Financial Crisis ob Child Mortality: The Case of Sub Saharan Africa. Innocenti Working Paper. 2011–01.

68.

Wamani

Åstrøm

Peterson

. (2007) Boys are more stunted than girls in Sub-Saharan Africa: A meta-analysis of 16 demographic and health surveys. BMC Pediatrics 7: 1–17.

69.

Webb

Reardon

(1992) Drought impact and household response in East and West Africa. Quarterly Journal of International Agriculture 31(3): 230–246.

70.

World Meteorological Organization (2013) The Global Climate 2001–2010. WMO No. 1119. UNDP.

71.

Yamano

Alderman

Christiaensen

(2005) Child growth, shocks and food aid in rural Ethiopia. American Journal of Agricultural Economics 87(2): 273–288.

72.

Yamauchi

Yohannes

Quisumbing

(2009) Natural disasters, self-insurance, and human capital investment: Evidence from Bangladesh, Ethiopia, and Malawi. IFPRI Discussion Paper 00881.