On the Origins of Confirmation Bias in Mainstream Analysis of Structural Change - What Constant Price Sectoral GDP Measures, What it Does Not, and Implications

The paper discusses the possible origins of what appears to be a striking example of confirmation bias in mainstream economics: the issue of industrialisation - why there has been general ignorance, so the data shows, of the observable main characteristic of structural change in developing countries since the end of the Cold War - servicisation, not industrialisation. The paper argues that an explanation for this can be found in the vulnerability of the standard analytical framework - the use of production functions - to consequences of the common unwarranted assumptions that sectoral ‘output’ is measured by published constant price sectoral GDP data, which are encouraged by core assumptions in its approach, which can be related to the assumption of instrumental rationality. Suggestions are made