Introduction: what the Levelling Up White Paper seeks to achieve

The Levelling Up White Paper (LUWP) published on 2 February 2022 makes the case for investment in six capitals (human, financial, social, physical, intangible and institutional), which it argues underpin economic and social development (HM Government, 2022). These capitals are inter-related: for example, financial capital is required for investment in other capitals. There are also capitals which are not included, for example natural or environmental capital, although the LUWP argues that these are picked up by other policies (HM Treasury, 2021; Dasgupta, 2021). This absence is surprising, especially considering the attention they were given in a publication on which the LUWP draws to develop its ‘capitals’ approach (Coyle et al., 2019). As the base upon which all other economic activity can proceed, our attention here is specifically with the role of physical capital in levelling up and its treatment by the White Paper. This paper focuses on one capital – physical capital – and, whatever the fate of levelling up as a policy agenda, why it is an important part of understanding and addressing regional inequality in the UK.

The LUWP includes an analysis of Gross Fixed Capital Formation (GFCF) and highlights that correcting regional imbalances could be a key element of levelling up. GFCF is an estimate of physical capital investment by all sectors (both public and private). It includes spending on plant and machinery, ICT equipment and software, as well as areas such as the construction and improvement of buildings, roads, railways, and underground infrastructures such as sewer and cable networks. Requirements for fixed capital will vary by sector in terms of scope and scale – for instance, a software company will require ICT investment whilst a petrochemical works will have a huge requirement for plant and machinery. How this capital is used can matter as much as how much capital there is; thus, more highly skilled workers may get more from the capital in terms of output than less skilled workers. The amount of physical capital may grow due to domestic investment but also from foreign direct investment (FDI). Physical capital reflects investment in assets which will bring long term benefits to an economy – in the case of large plant or buildings this may be over 50 years or more.

Surprisingly the LUWP says relatively little as to why physical capital is important to levelling up. In terms of the medium-term aspirations of the LUWP’s mission on productivity the focus is primarily on spreading research and development expenditure, investment in public transport outside London, and in providing high-speed broadband and 5G. Little is said about what might be seen as traditionally important parts of GFCF such as renewing and modernising the built environment and upgrading plant and machinery across different sectors; nor how physical capital is required to meet future economic drivers, such as automation. The approach the LUWP takes is to seek to facilitate investment through improving quasi-public financial support for SMEs (such as the British Business Bank) and by unlocking institutional investment, with the role of pension schemes (such as the Local Government Pension Scheme) highlighted as an opportunity to better align these funds with investment in projects that will have local and regional benefits.

Prima facie this seems like a limited approach. The next section explores these issues further and considers why fixed capital matters for levelling up and more broadly for economic and social development.

Why Physical Capital Matters

GFCF, as the main measure of physical capital, varies markedly between countries, in part reflecting their stage of development (for instance, rapidly industrialising countries would be expected to have higher GFCF than an already industrialised country). This emerges clearly from the trends shown in Figure 1. Hence, since 2000 China’s GFCF has on average been 40 per cent of GDP (and rising) – although in part this is due to considerable expansion of urban housing, Germany and France just over 20 per cent (and constant) and that of the UK around 17 per cent (and slowly falling, as well as lagging behind the USA for the last 30 years). The differences between the UK, France and the UK may be partly explained by sectoral mix – with Germany and France having larger manufacturing sectors and historically larger public programmes of investment in areas such as transport infrastructure. There is evidence that higher levels of physical capital investment lead to higher levels of growth (De Long and Summers, 1991), although it should be stressed that other factors are also important (Lipsey and Kravis, 1987), including investment in the other capitals.

Authors such as Blomström et al. (1996) have also raised questions around causation: for example, is it higher levels of physical capital investment which lead to growth or vice versa? As these authors conclude: ‘Institutions, economic and political climate, and economic policies that encourage education, inflows of direct investment, lower population growth, and the efficient use of investment seem to be the chief foundations for economic growth’ (ibid., 276). However, at the very least GFCF is a good indicator of the investment by which these policies and their attendant institutional frameworks lay the foundations for growth. As Rodríguez-Pose and Ketterer (2020) show, there may be differences between nations and regions in terms of investment needs. Thus, in southern EU regions improvements in governance and institutions are needed to make effective use of investment, whereas in central and eastern EU regions increases in total investment are required to compensate for the neglect of the late Warsaw Pact years.

What’s going on in the UK?

What is striking about the UK is that there has been a distinct spatial pattern of physical capital expenditure since at least 2000. Table 1 compares regional GFCF (our preferred measure of investment in physical capital) to the UK average between 2000 and 2018 and shows the ‘winners’ and ‘losers’ over this time, colour-coded into black and red type respectively. The pattern that emerges largely reflects a north-south divide. London, the South East and East of England have been running ahead of the UK average for the entirety of this period. At the same time there are a couple of outliers: Scotland which has consistently seen higher levels of GFCF for 20 years than all other regions bar London, although some evidence suggests that this is due to continued investment in North Sea oil and gas in North East Scotland/Aberdeen region (Gardiner et al., 2020); and Northern Ireland which appears to have benefited from a ’peace dividend’ for GFCF, at least during the first decade of the 2000s. Wales, the Midlands and the North of England have all lagged behind the South of England and Scotland. This is largely an extension of the analysis in the LUWP (p.76), although we have extended it by looking at different time periods. The disparities have persisted between the time of the New Labour (1997-2010) and the Coalition/Conservative (2010-present) governments, albeit on a much larger scale during the more recent years of austerity-based policies.

We have also examined investment differentials between countries and regions on a per capita basis (Table 2). This reveals similar overall patterns but also amplifies the scale of what appears to be the Northern Ireland peace dividend in the early 2000s, and also just how well London and Scotland have fared compared to the rest of the UK.

The two tables highlight the scale of the challenge for levelling up. From a path dependency perspective (Hassink et al., 2019), turning around these patterns of investment will be very challenging, even by the end of the decade. The analysis raises major questions around the extent to which such disparities might be addressed. Moreover, it raises the question of whether the focus should be on changing the conditions which promote greater GFCF (as Blomström et al. (1996) would argue), or whether it would be better to use subsidies and incentives to stimulate greater physical capital investment (e.g., in the industries identified in the UK’s previous Industrial Strategy) and adopt related measures to address poor productivity.

Research led by Ben Gardiner (Gardiner et al., 2020; 2021) has attempted to untangle the relationship between capital stocks and productivity at a regional level. It shows that capital intensity (capital stock per capita) does have a positive effect on labour productivity, especially in manufacturing sectors, where the authors point to the rising significance of automation. They also note that investment in business services, especially in information and communication technologies (ICT), may understate the significance of the actual role of investment in promoting labour productivity growth. In simple terms ICT investment can have a transformative effect on business activity, although evidence from the last 15 years suggests that the benefits of the ICT revolution have not yet been realised in higher levels of economic growth (van Ark et al., 2021). Unravelling the relationship between investment in capital and in skills is a key challenge in addressing the ‘productivity puzzle’ (Haldane, 2017) in the UK. In brief, this refers to the UK phenomenon where 20 to 25 per cent of firms are more productive than comparators in the same quintiles and quartiles in similar industrialised countries, whilst the remainder lag behind their international equivalents on the same indicator. This points to the risk of a ‘two speed’ economy, although Haldane is keen to stress that the most productive firms are found in all regions of the UK.

Why do regional variations persist?

It is not within the scope of this paper to explore in depth why such differences in regional GFCF might exist, but it is possible to draw a stylized account of some of the broad factors behind the divergence. Firstly, we are seeing a continual sectoral shift playing out in the UK with a decline in the proportionate share of manufacturing and a continued rise of services, with what are seen as productive service functions in finance and other professions concentrated in London and to some extent in cities such as Edinburgh. Secondly, as Table 3 shows, the older industrial regions received lower shares of public expenditure during the austerity years between 2010 and 2020, both overall and in terms of capital investment. At the same time London and surrounding regions benefitted from growing portions of what was effectively a smaller pie.  This was not just from welfare budgets and local authority spending but also in terms of the scale of regional development funding and the role of regional and local economic development infrastructures (Gray and Barford, 2018). The Local Enterprise Partnerships (LEPs) have been poor replacements in scale and scope to the Regional Development Agencies (RDAs), despite the rise in City Deal type funding.