02The geography of the jobs miracle

The 2.7 million new private sector roles equated to an increase of 13.7 per cent. Of those posts, 1.7 million were in cities and large towns that saw a higher private sector growth rate of 15 per cent. This section sets out how the jobs miracle played out across the country.6

Job creation was centred on London and other big cities

The capital accounted for 790,300 (29 per cent) of new roles – that’s 17 Scarboroughs or 25 Hartlepools. It means that more than one in four net private sector jobs created up to 2019 were London-based. Other big cities played an outsized role too, including Manchester (152,100 positions), Birmingham (99,100) and Glasgow (40,800).7

The 10 largest cities, including London, which make up just 3.5 per cent of all land, accounted for 45.6 per cent of private sector jobs growth between 2013 and 2019 (see Figure 1).

Figure 1: London led the jobs miracle, followed by other big cities

Source: ONS, Business Structure Database (BSD)

Box 1: What is job creation?

The analysis considers only private sector job creation as it is the performance of the private sector that determines the variation in economic performance across the country. This report uses the following terms:

Job gains – all roles created in a place between 2013 and 2019. This includes those that existed for a short period (e.g. one year) or a post that was added by a business starting up in 2013 that was retained until 2019.

Job losses – as with gains, this is all jobs lost between 2013 and 2019, including those that were lost one year and replaced the next.

Job churn – the sum of all job gains and losses divided by the number of jobs in 2013. This acts as a measure of dynamism, with more dynamic places seeing a greater number of posts created and destroyed.

Net job creation – the difference between gains and losses over the same period. If a place gains more than it loses, net job creation will be positive. If there are more losses, it will be negative. Net job creation does not say anything about the underlying churn and job dynamics of a place.

Looking at percentages rather than absolute growth, a number of smaller places performed well (see Figure 2). Oxford and Milton Keynes were the strongest, adding more than one extra job for every five that existed in 2013.

Three of the top five cities were in the Greater South East. However, strong performance was not limited to this region – with increases of more than 20 per cent, Burnley and Barnsley saw the third and fifth strongest growth, respectively. Only one city lost jobs overall. Aberdeen had 2 per cent fewer roles in 2019 than in 2013. In total, 22 cities and large towns had a private sector growth rate above the national rate, while the other 40 lagged behind.

Figure 2: Strong jobs growth was not limited to the Greater South East

Source: ONS, Business Structure Database (BSD)

Private sector net job creation, 2013-19


Percentage change

Absolute change

1 Oxford 25%                                 12,700
2 Milton Keynes 23%                                 27,000
3 Burnley 23%                                 10,500
4 Peterborough 22%                                 17,000
5 Barnsley 21%                                 10,200
6 Liverpool 20%                                 36,500
7 Reading 20%                                 25,700
8 London 20%                               790,300
9 Manchester 20%                               152,100
10 Northampton 19%                                 17,000
53 Norwich 6%                                    6,200
54 Middlesbrough 6%                                    7,200
55 York 6%                                    4,100
56 Preston 6%                                    7,000
57 Worthing 6%                                    1,500
58 Portsmouth 4%                                    5,900
59 Edinburgh 3%                                    6,300
60 Wigan 3%                                    2,000
61 Nottingham 2%                                    4,300
62 Aberdeen -2% – 2,200
GB 14%                            2,729,500

Source: ONS, Business Structure Database (BSD)

Net job creation hides a much larger churn

Job growth figures mask the far greater numbers of positions gained and lost. At a national level, to achieve more than 2.7 million new net private sector roles, which is what happened during the jobs miracle, more than 19.3 million were created in the six years from 2013. Alongside this, approximately 16.6 million were lost. This meant around seven had to be generated to produce one net private sector job.

If this pattern were to hold post-pandemic, Great Britain would need to create almost 9.4 million jobs over the next few years to provide roles for approximately 1.3 million people who are out of work because of Covid-19.8 And this is before considering the longer-term challenges of levelling up.

Figure 3 shows these job gains and losses for each of the 62 largest cities and towns in Britain. Three important points can be drawn from it:

  • There is a negative relationship between jobs created and lost in a place. Cities and large towns that generated few roles also lost few, whereas those that produced many jobs saw more lost. No place was able to create a lot of jobs while only losing a few.
  • Cities and large towns with similar rates of growth achieved it in different ways. It is most starkly shown in the contrast between London, Britain’s most dynamic over the period, and Burnley, which was one of the least. Figure 2 shows both places are in the top 10 for private sector jobs growth (also indicated in Figure 3 by how far above the ‘break even’ line they sit). London achieved this with very high churn, managing to keep job creation above its high rates of losses. In contrast, Burnley kept losses down.

Meanwhile, Aberdeen’s net fall in private jobs was not because it struggled to create roles – it generated more per 10,000 private sector jobs than 24 other cities and large towns. The problem was that it lost them faster than it created them.

Theirs is only a weak link between dynamism and economic performance, despite evidence from the US suggesting a stronger one.9 On average, cities and large towns were more dynamic than non-urban areas. Most that were less dynamic than the non-urban rate, such as Telford, Preston and Stoke, tend to have low wages, and no city or large town in the bottom 30 for wages had levels of dynamism above the urban rate. But there are a number of high-wage cities and large towns, such as Derby, Cambridge and Slough, that do not stand out as being particularly dynamic. It is not clear why this is the case.

Figure 3: The most dynamic cities have high levels of job creation and loss

Source: ONS, Business Structure Database (BSD), ONS, Annual Survey of Hours and Earnings (ASHE), average gross weekly workplace-based earnings. Own calculations for PUA-level weighted by number of jobs, CPI inflation adjusted.

What is clear is that high levels of churn are not a bad thing. Job losses are not necessarily a concern from the viewpoint of a city economy, so long as the city can generate new roles to replace them.


  • 6 This project only looks at Great Britain rather than the UK, as data for Northern Ireland is not available for all indicators in the Business Structure Database.
  • 7 Those with a population over 600,000 in 2019. These are London, Birmingham, Manchester, Glasgow, Newcastle, Sheffield, Leeds, Nottingham, Liverpool and Bristol.
  • 8 Centre for Cities (2021), Cities Outlook 2021, London: Centre for Cities
  • 9 Haltiwanger J (2015), Job creation, job destruction, and productivity growth: The role of young businesses, Annual Review of Economics, 7(1): 341–358; Faberman R (2005), What’s in a city?: Understanding the micro-level employer dynamics underlying urban growth, Washington: U.S. Bureau of Labor Statistics