The main objective of our analysis is to investigate the causes of shifts of the Beveridge curve in Great Britain, the Netherlands and Sweden. In chapter 2., we will outline the model which is the basis for our analysis. The cornerstone of our theoretical framework regarding the long-run relationship between unemployment and vacancies otherwise known as the Beveridge curve is the matching process. We will describe how certain features such as structural mismatch, the relative attractiveness of benefit provisions and changes in search intensity and search effectiveness of the unemployed, could theoretically affect the Beveridge curve. In order to analyse a possible shift of the Beveridge curve, time series analysis will be used. In chapter 3., we describe the patterns of the relevant data series. Also, we describe the significance of the long-term unemployment problem in Great Britain and the Netherlands. In the following chapter we describe the characteristics of the disability arrangements in the three relevant countries. We do this in order to explain how the disability arrangements have affected the unemployment patterns in one of our sample countries. The focal point is the existence of a hidden unemployment component in the disability stock. In order to estimate the Beveridge curve for each country, we will use the instrumental variables technique. In chapter 5., after first having tested for the suitability of our econometric practice regarding the data series in the context of the theory of cointegration, we will present and discuss several model specifications regarding the Beveridge curve. We will also test for the sensitivy of our main results to variations in data and estimation method. Also, we present models of the British and Dutch long-term unemployment patterns. In chapter 6., we will discuss the most relevant results and compare the British, Dutch and Swedish labour market experiences. Conclusions are presented in the final chapter.