Child poverty analysis
• A significant amount of statistical information on the wellbeing of children is already available for each of the three countries, although there are gaps that can be filled in the longer term.
• Most importantly, available data on disparities in the wellbeing of children are significantly under-analysed. An exception is the 2008 UNDP report for the Solomon Islands on the estimation of basic needs poverty lines, and the
incidence and characteristics of poverty in Solomon Islands. This does contain information on child poverty, but even in this case the analysis is relatively limited.
• It is recommended that new in-depth analysis should be undertaken of the circumstances of children in the three countries. The primary data sources for this analysis should be the HIES for each country, plus the DHS for the Solomon Islands and the MICS for Vanuatu. As far as possible, this analysis should attempt to complete the tables required for the Global Study of Child
Disparities.
• In order to deepen and strengthen this analysis it is recommended that a wider range of alternative measures of child poverty be prepared. Due to the lack of estimates of Purchasing Power Parities (PPPS) for these countries it is not currently possible to estimate proportions of the population living below the MDG relevant poverty lines.
• The UNDP (2008) has estimated Food Poverty Lines and Basic Needs Poverty Lines for the Solomon Islands for 2005-06. The same methodology could be applied to estimate Food Poverty Lines for Kiribati and Vanuatu.
• It is also useful to estimate a range of alternative poverty lines based on relative measures of poverty, for example, 40%, 505 and 60% of median household expenditure.
• In addition, it would be important to take account of the impact of alternative estimates of equivalence scales for households of different size and composition.
• International best practice suggests that using a range of poverty lines and equivalence scales is the most appropriate way of “triangulating” results. In particular, if different methodologies produce different results in terms of estimated levels of poverty or in particular the profile of groups vulnerable to poverty then caution is needed in drawing strong policy conclusion. However, if different methods produce similar or at least consistent results then these results are likely to be more statistically robust and reliable for policy purposes.

Evaluating progress towards the Millennium Development Goals
• There is a significant international literature evaluating progress towards the Millennium Development Goals in other parts of the world outside the Pacific. A range of methodologies have therefore been developed, but almost no related studies have been undertaken of Pacific countries.
• The next step therefore is to replicate the methodologies used in a range of these studies on available data from Kiribati, the Solomon Islands and Vanuatu.
• Three methodologies are recommended.

o The first is to undertake time series analysis of trends in the main childrelevant MDGs since 1990 and relate these trends statistically and descriptively to economic and policy trends in each country.
o The second type of approach is to look at cross sectional data on infant and child mortality and factors influencing these outcomes. The most appropriate data sources for such an analysis are the MICS survey in Vanuatu and the DHS in Solomon Islands. Suitable data are lacking for Kiribati. In the first instance, this analysis would be primarily descriptive in the sense that the analysis would be designed to identify what factors within each country appear to be related to variations in MDG indicators.
o The third recommended area of analysis is to compare MDG outcomes across Pacific Island countries identifying leaders and laggards in progress towards these goals, and then on the basis of a review of the literature on economic and social trends in Pacific countries attempt to identify possible explanations for these variations in outcomes. As well as the three countries under review, it is recommended that analysis be undertaken of the best-performing and the worst-performing countries in the Pacific as a means of maximising variability and identifying possible policy effectiveness.