One of the most important development goals is the reduction in absolute poverty to 15 per cent by 2015. This and related development goals have been agreed upon by governments and the UN system, and have been labelled the Millennium Development Goals. In my recently published book, Imagine There's No Country: Poverty, Inequality and Growth in the Era of Globalisation, I had documented how the poverty reduction goal had already been reached by 2000, the very year of formulation of the goals for 2015. In a critique of my study, World Bank as well as its main poverty analyst, Martin Ravallion, question the authenticity of the data, assumptions and methods used by Imagine. In fact, data and definitions account for an insignificant amount of the difference in the poverty estimates of World Bank and Imagine. The major explanation for the higher World Bank poverty rates is found to be due to a lower growth estimate of per capita expenditures, and especially lower compared to the growth estimate obtained from national accounts data. This lower growth, 10.4 per cent over 11 years, 1987-1998, is based on household survey means (World Bank data). An associated, and surprising, finding is that while poverty estimates are accurately reproduced, there is a big divergence between the published growth rate of 10.4 per cent and the 'reproduced' survey growth of 5.6 per cent. Notwithstanding this major uncertainty about the World Bank data or its growth and poverty results, all the major findings of Imagine are faithfully reproduced exclusively using only World Bank data. Further, an extension of the World Bank poverty measurement method also yields the result that the MDG poverty reduction goal has been reached. Finally, using the recently released 1996 PPP data, poverty in 2000 was below 15 per cent for all methods, including the flawed World Bank poverty measurement method.