The matching method for treatment evaluation does not balance selective unobserved differences between treated and non-treated. We derive a simple correction term if there is an instrument that shifts the treatment probability to zero in specific cases. Within the same framework we also suggest...

The matching method for treatment evaluation does not balance selective unobserved differences between treated and non-treated. We derive a simple correction term if there is an instrument that shifts the treatment probability to zero in specific cases. Policies with eligibility restrictions,...

This paper considers the issue of unemployment one of the most pressing issues facing the UK and other governments, as the current recessions deepens. It begins by trying to accurately date the beginning of the current downturn in the British economy, arguing that it is clear that the recession...