Uk Labour Market Reforms and Sectoral Wage Formation

This paper estimates the impact of UK labour market reforms on UK wages in seven industrial sectors. We estimate real wage equations, disaggregated by industry, and find that the introduction of labour market reforms corresponds with an increase in wage flexibility for some of our sample industries from 1986 onwards. The main findings of this study are as follows: 1. Unemployment only has a significant impact on wages in sectors which have a high-proportion of less-skilled workers whose pay is particularly low (such as construction and distribution and repairs). It therefore seems that the threat of unemployment puts more downward pressure on the wages of the less-skilled as they are more likely to become unemployed and have a smaller probability of finding another job relative to skilled workers. Hence the less-skilled and the lowpaid seem to account for the bulk of any wage flexibility evident at the macro-level. 2. Because the long-term unemployed gradually become disillusioned, and their skills become eroded, they tend to exert less downward pressure on wages compared to the short-term unemployed (as they become less active or effective in searching for work than those who lost their job more recently). However, we find that UK labour market reforms such as ‘Restart’ interviews may have increased the downward wage pressure exerted by the long-term unemployed from 1986 onwards. Although the reforms may have pushed the long-term unemployed into attending more job interviews for ‘hard-to-fill’ vacancies, there is little evidence that government schemes helped them compete more effectively for jobs by enhancing their skills, etc. It is more likely to be the case that ‘Restart’ pressurised the long-term unemployed into taking lower-paid jobs than they otherwise would (eg, we only find extra downward pressure exerted by the long-term unemployed in low-paid sectors which is consistent with other evidence that jobs being taken by the unemployed typically offer substantially lower wages than other jobs). Another possibility is that policies such as ‘Restart’ pushed those not actively seeking work out of unemployment into ‘inactivity’ hence the long-term unemployed now primarily consist of those engaged in active job search. However, this latter explanation may imply that there has been no change in the downward wage pressure exerted by the total number of individuals who have been without work for more than twelve months. 3. Some sectors which indicate that the labour market reforms increased the downward wage pressure exerted by a given rise in unemployment also show a positive and significant intercept dummy which suggests an unexplained rise in wages. We argue that this is consistent with the considerable increase in the wages of skilled workers relative to the less-skilled in the UK since the early 1980s. One explanation may be that the positive intercept dummy represents the rapid rise in wages for skilled workers (which, in turn, represents a decline in the probability of unemployment for the skilled) and the increase in the unemployment parameter captures the considerably slower wage growth for the less-skilled due to a rise in the probability of unemployment for workers in this category (which is consistent with evidence that labour turnover has increased markedly for less-skilled males). Skilled and less-skilled workers may have experienced opposite movements in the threat/probability of unemployment in recent years because: first, a rapid-rise in skill-biased technical progress seems to have increased the demand for skilled workers relative to the less-skilled; second, the increasing use of temporary and parttime workers may create a ‘buffer’ labour force who are more likely to lose their jobs than the ‘insiders’ (ie, the skilled, permanent, workers in the firm) hence reducing the threat of unemployment to the skilled and thereby increasing their bargaining power. a I am grateful to the Leverhulme Trust for providing finance for the research contained in this paper.