there is approximately 0.02 reduction in vulnerability (see Figure 13). This reduction becomes slower in higher positive changes in temperature, about 0.01. Changes in maximum temperature demonstrates effects contrary to minimum. A degree higher change in maximum temperature effects about 0.02 drop in log crop farming income. On higher changes in maximum temperatures (0 degrees above), the drop in crop farming income is less by .01. These effects in turn translates to approximately 0.002 increase in vulnerability, although the rate becomes slower by 0.002 in higher changes. Figure 13. Effect of change in minimum and maximum temperature on vulnerability, CBMS 20072010 Covariates Demographic characteristics Age of the household head is consistently associated with lower probability of being malnourished across all climate shocks. In addition female-headed households are less likely to have malnourished children. On the other hand, households with indigenous member are more likely to have malnourished children as well as those who have experienced illness in their households. Higher educational attainment of the household head appear to reduce probability of malnourished children. Household assets and infrastructure Ownership of assets and access to infrastructure have mitigating effects on vulnerability to food insecurity in terms of malnutrition. Those with safe water are less likely to have malnourished children compared to those without safe water. Access to communication such as cellphone and telephone also reduces the likelihood of having malnourished children. This is also similar with those who have access to media and information such as radio. On the other hand, access to government programs is likely associated with malnutrition as well as higher prices of food. Policy Variables Agricultural assets shows a robust effect on agricultural income and in turn on food insecurity. For instance those with agricultural organization, thresher, dryer or irrigation pump will likely have more income on CBMS-FAO | 41