Background: There are large variations in vaccination coverage, not only between high and low-income countries but also across low income countries and within low income countries. The reasons behind these variations are only sketchily understood. In particular, the current understanding of demand for childhood vaccinations is limited. Due to inadequate vaccination coverage more than one million children die annually from vaccine preventable diseases. Objective: The study set out to examine demand for childhood vaccinations from an economic perspective: to identify caretakers` perceptions of potential costs and benefits of vaccinating a child, and to examine the association between these perceptions and caretakers` decision making for childhood vaccination. Furthermore the study seeks to identify variables associated with caretakers` perception of benefits of vaccinating a child. Methods: The study was cross sectional, used structured questionnaires and employed a two stage cluster sampling technique. Respondents were caretakers of children at the age 18 - 59 months, in total 635 respondents were included in the study. The study was conducted in two traditional authorities in Thyolo district, Malawi. Descriptive statistics were used to describe the variables of study. Logistic regression analyses (univariate and multivariate) were conducted to measure the association between predicted explanatory variables from economic theory and decision making for childhood vaccination, and to examine the relationship between predicted explanatory variables and perceived benefits. Results: 96.1 percent of the respondents reported to fully have vaccinated their youngest child in the age 18 – 59 months for all routine EPI vaccinations. The large majority of caretakers scored the measured benefits of vaccinating a child to be high, while they to a large extent were divided in their perceptions of costs. A large share of caretakers had to travel substantial distances to vaccinate their children. Incorrect knowledge of vaccination schedule (OR = 2.95 (CI 0.97 – 8.99) P= 0, 06), fear for severity of side effects (OR= 3.8 (CI 0.89-16.17) P= 0.07), distrust in information on vaccination (OR=27.55 (CI 5- 149) P=0, 00) and giving birth at home (OR=2.52 (CI = 1.18-5.39) P=0.02) were found to be determinants for vaccination default (not having fully vaccinated youngest eligible child for all EPI routine vaccinations) in the univariate analysis. Not any of these determinants remained significant in the multivariate regression analysis (p-value < 0.05). Distrust in received information (OR= 27.52 CI (6 – 131) P=0.00) and being aware of less than two side effects (OR= 2.32 (CI 1.15- 4.68) P=0.019) were found to be determinants for limited perceived benefits (scoring the preventive effect of vaccination as limited) in the multivariate analysis.Discussion and conclusion: The study documents and points to the possibility and necessity of achieving high vaccination coverage in areas where many caretakers need to travel long distances to reach vaccinations, and where a large number of caretakers perceive the traveling and waiting time as long. The study suggests that high level of trust in information and in vaccinators may be an essential explanatory factor; in the way that trust facilitates positive perceived benefits which again make caretakers seek childhood vaccinations even though there are considerable costs involved. The study, however, does not provide the final explanation for why caretakers in the study area vaccinate their children, and nevertheless for why caretakers vaccinate or do not vaccinate their children in other areas. More emphasis should be devoted to demand for childhood vaccinations, both in research and in policy making.