Objectives:  Cost-effectiveness Analysis (CEA) and the calculation of the Incremental Cost-Effectiveness Ratio (ICER) together with its comparison with a threshold such as Gross Domestic Product (GDP)/capita, have long been used to assess the value for money of a new intervention compared with a comparator that this new intervention precisely seeks to displace.  In this paper we show the paradoxical increase in cost-effective price using data from middle, low and very low income settings. Methods:  Using the introduction of rotavirus vaccination compared with no-vaccination as the example.  We create a theoretical framework for calculating the ICER by gradually decreasing the investment for treatment of rotavirus related disease (the ‘no-vaccination comparator’) representing different countries with different GDP levels and decreasing levels of existing healthcare investment. We compare these results with an analysis of cost-effectiveness using real data from 9 countries representing a range of different GDP levels. Results:  The theoretical framework works well in situations where the GDP/capita exceeds $10,000 – as expected the cost-effective price decreases with a decrease in the GDP/capita.  Below this the scant investment in healthcare infrastructure, thereby reducing potential cost-offsets, coupled with the significant increase in the potential effect gain, results in a much wider margin between a cost-neutral and cost-effective price that could effectively be set using this approach.  Conclusions:  Although Cost-Effectiveness Analysis is widely used to assess the value for money of a new intervention for a particular price, we would argue that where investment in healthcare is low and disease burden is high, the use of CEA leads to paradoxes in price-setting.