Policy and Politics: To tackle low pay, policymakers must think about sectors

Policy and Politics: To tackle low pay, policymakers must think about sectors

Neil Lee

This section of Discover Society is provided in collaboration with the journal, Policy and Politics. It is curated by Sarah Brown.

The UK has a low pay problem. For some time, UK labour market policy has focused on the supply side of the labour market. The aim has been to get people into work, with little concern for the type of job they take. Low pay has been addressed using a minimum wage, while economic development policies have focused on high-end sectors and investments in science and innovation.

The result has been success on one indicator, with relatively high employment rates, but a failure on another, with one of the highest levels of low pay in the OECD. Around a fifth of the UK’s workforce are in low pay, defined as earning less than two thirds of national median weekly pay. Low pay has been compounded by wage stagnation: between 2008 and 2015 the only OECD country with worse wage growth was Greece.

Much of this low pay is in a small number of sectors, some of which are forecast to grow significantly in the future, worsening the low pay problem. Through our close analysis of sectoral differences and the dynamics of low-paid sectors in our new article in Policy & Politics, we reveal that instead of the current policy focus, efforts to improve productivity and earnings mobility in low-pay sectors, could improve living standards as well as the UK’s overall economic performance.

Low paid jobs exist in all sectors of the economy. Table 1 shows the share of workers in each sector in low pay, how much of the labour force each sector represents, and the share of total low paid employment each accounts for. Even in the sector with the lowest rates of low pay, Public Administration and Defence, almost 4 % of workers earn below the low pay threshold. The sectors with low rates of low pay often have high-skilled workforces (e.g. Finance and Insurance where only 5% are low paid) or are highly unionised (Electricity, gas and water supply where 4% are).

The three sectors with the highest share of low pay – Accommodation and Food Services, Residential Care, and Retail and Wholesale – have rates of low pay of around 59%, 40% and 39% respectively. Taken together, these sectors account for almost half of the UK’s low paid workers. They tend to have low levels of productivity: waiting at tables work cannot (yet) be done by a robot, and shop-work is still a large employer on relatively low wages. Residential care is an outlier, as much is funded by the state through local authorities, although it is often outsourced to the cheapest provider, meaning wages are squeezed. Importantly, working in one of these low-paid sectors also makes it harder to escape low pay – they tend to have fewer upward wage moves.

Table 1. Low pay by sector (Labour Force Survey, 2010-2014)

Sector Low pay

(% workers in sector in low pay)

Size of sector

(% of total employment)

Low pay share

(% of all low paid employment)

Agriculture 35.4 0.6 1.0
Mining 4.5 0.4 0.1
Manufacturing 14.8 10.7 7.7
Electricity, gas, and water supply 4.0 0.6 0.1
Water supply, sewerage and waste 11.5 0.8 0.5
Construction 12.9 5.2 3.3
Wholesale and retail 39.1 14.5 27.4
Transport and communications 14.4 4.8 3.3
Accommodation and food services 59.1 5.7 15.8
Information and communications 7.2 3.7 1.3
Financial and insurance 5.0 4.4 1.1
Real estate activities 10.9 1.0 0.5
Professional, scientific and technical activities 8.6 5.7 2.3
Admin and support services 29.7 4.4 6.2
Public administration and defence 3.8 7.2 1.3
Education 16.6 11.3 9.0
Arts, entertainment and recreation 29.2 2.2 3.1
Other service activities 34.0 2.2 3.3
Human Health 8.2 7.8 3.1
Residential care 40.0 3.4 6.3
Social work 19.5 3.3 3.2
Total 20.8 100.00 100.00


These sectors have a number of characteristics, not least their reliance on low skilled workers, so one objection is that these are simply the result of low-skilled workers sorting into low-paid sectors. Yet our results suggest that this isn’t the case. We run a series of statistical models, in which we control for the impact of the other personal characteristics, including education levels, on the probability of a worker being in low pay. Figure 1 gives the results. These show a clear independent sector effect on low pay. A worker in Wholesale and Retail is more than 50% more likely to be low paid than one in Manufacturing, even controlling for their other characteristics. A worker in Accommodation and Food Services would be around 80% more likely to be in low pay than an otherwise identical one in Manufacturing.

Figure 1. Impact of sector on probability of low pay, controlling for personal characteristics, UK 2010-14

Source: Labour Force Survey. Dots indicate the marginal effect for each sectoral variable of the regression presented in table 5, column 2. Lines give 95% confidence intervals, with red bar indicating zero. Reference category = Manufacturing. Controls: education, age, age2, country of birth, region, and year/quarter fixed effects.


Future labour market trends show that there will be significant growth in some low-paid sub-sectors. Working Futures projections for future employment suggest that by 2024 there will be another 190,000 low paid workers in Accommodation and Food Services and 131,000 in Wholesale and Retail. Efforts to ensure ‘good jobs’ are created in these sectors will be important to help address the UK’s poor wage performance. The only low-pay sector predicted to decline significantly is Agriculture and Manufacturing, which still offers some well-paid jobs for those without degree qualifications, will decline relatively more.

Figure 2. Low pay and projected employment growth, 2014-2024

Source: Labour Force Survey + Working Futures 2016.


What does this mean for policy? The dominant policy approach has been on the supply side of the labour market, with little effort to shape the sectoral composition of the economy. Yet the idea that policymakers can shape the demand side of the labour market is now back on the agenda. The recent Industrial Strategy White Paper suggested that “Some of the biggest opportunities for raising productivity come in sectors of the economy that have lower average productivity levels, but where many people work”. These are the same sectors in which many workers earn low pay, and in tackling the productivity challenge in these sectors policymakers can help achieve an important dual goal – increasing productivity – and therefore economic growth – while reducing low pay – and increasing living standards. Getting as many people into employment as possible, regardless of the job, has been successful in some ways, but we also need a policy agenda to address the problems of this approach. The challenge will be reducing the extent of low pay, but to ensure the employment rate remains high. For example, automation and e-commerce may increase productivity in retail, but may reduce employment. We still don’t know how to balance these two goals.


Neil Lee is Associate Professor in economic geography in the Department of Geography and Environment at LSE.

Image: @eddileespinal

1 Comment responses

  1. Avatar
    February 15, 2018

    This makes me wonder about two things, a) what causes low pay/ low productivity in those sectors (and, related, what does low productivity mean) b) why do a proportion of highly educated/ skilled take low pay (and who are they).
    Having a degree from an elite university, I worked for low pay in retail for years. One, I was the wrong kind of graduate, without the social and cultural capitals that increase the probability of moving into a ‘good job’. I also felt a need to be in London, so settled for a retail job as that would let me pay the rent: I couldn’t afford to do an internship to get the experience I needed. With a steady supply of graduates in a similar position, retail companies know they don’t need to pay more. Indeed, if they paid more, they wouldn’t have to raise prices by much, so we could have higher pay/ productivity if society was willing to pay only a little more for its stuff, and have less stuff.
    Of course, many of those doing this eventually manage to get out of it, becoming authors, academics, and lawyers between 10-20 years later. The problem is, unlike the low pay/ high-experience-value you get in a first arts job, what I didn’t realise was that my years in retail would not count for anything later. So it’s easy to get stuck.


Leave a comment