Mark Williams looks at how occupations relate to the massive rise in British wage inequality between the mid-1970s and mid-1990s, finding that growing inequality is largely between groups and is driven by a small handful of occupations.
That Britain has become hugely unequal over the last few decades is a well-known fact. How inequality has grown is less well known. My research examines how occupations relate to trends in wage inequality. I found that the single-most important factor accounting for growing inequality was that the wages of already high-paying occupations increased the most, ahead of the changing relative sizes of occupations, and inequality within occupations.
Looking at what occupations people do could provide a promising avenue for policy makers, the general public, and academic research in understanding how the massive rise in wage inequality unfolded – and continues to unfold.
The LSE economist, Henry Phelps Brown, once remarked that
“the pay of any two persons may differ for many reasons […] but the most conspicuous reason lies in differences of occupation. It is these that come first to mind as the source of systematic differences in pay” (1977:28).
Occupations as a unit of analysis since fell away in wage inequality research – so little is known about how occupations relate to the massive rise in British wage inequality between the mid-70s and mid-90s. This is what I investigated using a classification system of 366 occupational groups.
How can occupations shape trends in wage inequality? Occupations can affect inequality in three ways:
- Changes in the relative sizes of occupations – there could be a growth in high-paying occupations and low-paying occupations, with a simultaneous fall in middle-paying occupations, polarising the wage structure;
- Changes in average wages of occupations – the distance between occupational wages could grow – the wages of some occupations may grow very steeply – some may even fall;
- Changes in inequality within occupations – a final way is the distance in wages between individuals could increase within occupations increasing overall inequality.
Examining the relative weight of each of these mechanisms in accounting for trends in British wage inequality using a statistical technique, I found that changes in occupational mean wages to be the most important factor, accounting for 48.3 per cent of the growth in male wage inequality 1975-1996, and 45.7 per cent of the growth in female wage inequality 1975-1996. Changes in the relative sizes of occupations is next important, accounting for 42.9 per cent for men, and a quarter for women. Least important is growing inequality within occupations, accounting for less than 10 per cent for men, and about a quarter for women.
The methodological approach I take allows for an estimation of the contribution by each and every single of the 366 occupations in the data. This is done by estimating a ‘counterfactual’ level of wage inequality – in effect calculating what the level of inequality ‘would be’ had occupation X’s size, mean wage, and internal inequality not changed – whilst all other occupations’ had changed. Subtracting this counterfactual level of inequality from the actual level of inequality reveals the contribution to the change in overall inequality of occupation X.
Some surprising results emerge. Just 10 occupations account for over half the growth in overall male wage inequality 1975-1996. The top 20 combined account for over two-thirds of the growth in overall inequality and cover just one-quarter of the 1996 labour force (16 per cent of the 1975 labour force).
For women, a similar finding emerges. Just 12 occupations account for over half the growth in overall inequality, and cover less than one-third of the 1996 labour force. The top 20 combined account for nearly two-thirds of the growth in overall inequality and cover just over 40 per cent of the 1996 labour force (and 33 per cent of the 1975 labour force).
For men, most are managerial, including the top four – with marketing and sales managers being the most important – accounting for almost 10 per cent of the growth in wage inequality alone. Not all the inequality-producing occupations are highly-paid, however, such as goods vehicle drivers, kitchen porters and cleaners. This is especially true for women where cleaners are the most important occupation. So when we think about growing inequality we should also think about what is happening amongst low-paying service occupations as well as high-paying managerial occupations.
As Phelps Brown noted, occupation is the first reason to come to mind when considering pay inequalities. This is because occupations relate to skills and aspect of employment relations such as span of control or position within the organisational hierarchy.
For policy-makers, the fact that growing inequality is largely between groups should be good news – as it gives them something more tractable to work with than if inequality is largely within groups. Most importantly, my research finds that growing inequality is driven by only a handful of occupations. For researchers, these results point to the need to pin down exactly what explains the ‘occupation effect’ and reasons why these particular occupations were as influential as they were.
My research adds to the wage inequality story by examining the role of detailed occupational categories – which have been oddly overlooked. My research has established that a between-group story is the more accurate one when groups are defined in terms of detailed occupations. My research points to the highly concentrated nature of growing wage inequality and to particular sections of the labour market that warrant further investigation. Moreover, the between-occupation story gives policy-makers something tangible to work with if tempering growing inequality is to be a policy goal.
Note: This article gives the views of the author, and not the position of the British Politics and Policy blog, nor of the London School of Economics. Please read our comments policy before posting.
Dr Mark Williams is a Fellow in the Employment Relations and Organisational Behaviour Group at the LSE. He recently completed his DPhil at Nuffield College, Oxford, where he also obtained an MSc in Sociology (Distinction). In addition to doctoral work he has conducted several other pieces of research including a government report on the role of workplace practices on mediating the quality of employment and an article on how country-level employment regimes shape the vulnerability of the low-skilled across Europe.