Finance and the Balassa-Samuelson hypothesis

Using financial data this dissertation attempts to investigate the Balassa-Samuelson hypothesis for a larger subset of countries and at a higher level of data frequency then previously examined using co-integration methods. The results are in-line with previous research, the real exchange rate does violate Purchasing Power Parity, and there is limited support for the Balassa-Samuelson hypothesis. This dissertation shows that a finance-based approach is a good proxy for the standard method usually adopted within the literature when investigating the Balassa-Samuelson hypothesis and also researchers should not be overly concerned with time-varying weights in the construction of price indices; finally as the frequency of data increases the Balassa-Samuelson hypothesis increasingly fails to explain movements in the real exchange rate.