Supply and spare capacity

The pace at which demand can grow without generating sustained inflationary pressures depends on the amount of spare capacity in the economy and the growth rate of potential supply. In turn, potential supply growth depends on structural features of the economy such as population growth and gains in productivity.

During the financial crisis, demand fell sharply, unemployment rose and a significant degree of spare capacity opened up. Potential supply growth also slowed as productivity growth stalled (Chart 3.1). In the years that followed, spare capacity was absorbed as demand grew faster than potential supply. The unemployment rate, for example, fell from 8% in 2013 to 4% by mid-2018 as companies increased hiring and reduced redundancies.

The MPC conducted its regular reassessment of supply-side conditions in the run-up to this Report. Spare capacity was judged to have been absorbed in 2018 Q4 (Section 3.1). Subdued demand growth over much of 2019 (Section 2), however, means that a degree of spare capacity is expected to emerge in the near term. The MPC judges that potential supply growth is likely to be slightly weaker than previously anticipated, at a little below 1½% in the central projection (Section 3.2).

The outlook for potential supply growth will be highly sensitive to the nature of the UK’s withdrawal from the EU. As described in Box 4 of the November 2018 Report, reductions in openness as the UK’s trading relationship with the EU changes are likely to reduce the economy’s productive capacity for a period of time. While such changes in supply could be relatively gradual in the event of a smooth withdrawal, a disorderly exit could severely impair the productive capacity of UK businesses.¹

• Notes

  ¹For further details, see EU withdrawal scenarios and monetary and financial stability: a response to the House of Commons Treasury Committee Opens in a new window.

3.1 Spare capacity in the economy

The degree of spare capacity is an important determinant of inflationary pressures. When resources are underutilised — for example if many people are out of work or if companies have spare productive capacity — there tends to be scope for output to rise without generating excess inflationary pressures. But once spare capacity is absorbed, rises in demand tend to put greater upward pressure on wage growth and inflation as companies need to pay more to recruit and retain staff or invest in additional resources.

The MPC uses a range of approaches to estimate spare capacity. One ‘top-down’ approach is to use statistical filtering techniques to estimate spare capacity from past observations of GDP, taking into account indicators of domestic price pressures. Weakness in core services CPI inflation, one measure of domestic inflationary pressures (Section 4), suggests that there might be a small margin of spare capacity in the economy. Results from filtering techniques tend to be very sensitive to the precise modelling assumptions, however.

An alternative ‘bottom-up’ approach is to separately estimate the components of spare capacity within the labour market and within firms. The unemployment rate — a key component of labour market slack — was 4% in the three months to November (Chart 3.2). That is a little below the MPC’s assessment of the ‘equilibrium rate’ of unemployment consistent with inflation at the target (Section 3.2).

A range of other indicators are also consistent with tight labour market conditions. Survey indicators of recruitment difficulties are above their past averages and the ratio of redundancies to employees is close to its historical low (Table 3.A). While most surveys of employment intentions softened a little in Q4, consistent with a slight slowing in employment growth in early 2019, labour market conditions are projected to remain tight and unemployment is expected to be broadly stable in the near term (Chart 3.2).

Spare capacity also appears to have been largely absorbed elsewhere in the labour market. The ‘marginal attachment’ ratio — the proportion of the working-age population who are not currently in work or seeking employment but report that they would like a job — has fallen sharply in recent years to a record low (Chart 3.3). That suggests the scope for further rises in the employment rate as such people enter the labour market is likely to be limited. In addition, the number of hours that those in employment say they would like to work, over and above those they are currently working, has fallen back in recent years and has been close to, or a little above, zero in recent quarters.

A further component of spare capacity is the extent to which companies’ capital equipment, such as vehicles or computers, is being underutilised. Reports from the Agents suggest that companies are operating at a little below normal capacity, meaning that firms may have some scope to raise output with existing resources (Chart 3.4). Results from the CBI and BCC surveys, however, suggest that spare capacity within companies has been absorbed.

Based on the evidence from both the ‘top-down’ and ‘bottom-up’ approaches, the MPC judges that demand and supply were broadly in balance in 2018 Q4. Spare capacity is projected to emerge in early 2019 as demand growth remains subdued (Section 2). Further out, inflationary pressures are projected to build as demand growth picks up while potential supply growth remains modest (Chart 3.1).

3.2 Potential supply

The supply capacity of the economy is determined by the quantity of labour available and the amount of output that workers can produce. Potential supply growth has slowed sharply since before the financial crisis due to persistent weakness in productivity growth (Table 3.C). While that weakness has been partly offset by robust growth in labour supply, overall potential supply growth has been around half its pre-crisis rate in recent years.

Following its regular reassessment of supply-side conditions, the MPC judges that potential supply growth will remain lower than its pre-crisis average, at a little below 1½% per year on average. That is slightly slower than projected in the November Report.

Labour supply

Population growth is a key driver of growth in labour supply (Table 3.C). Much of the UK’s population expansion over the past decade has reflected net inward migration, which peaked at over 300,000 per year in 2015–16 (Chart 3.5). While net migration remains higher than over much of the past decade, it has slowed since mid-2016 and was 273,000 in the year to 2018 Q2. That slowing was more than accounted for by lower migration from the EU; net migration from outside the EU rose to its highest level in over a decade.

The MPC’s forecast is conditioned on the ONS’s principal population projection, published in 2017. The projection implies a further slowing in net migration over the next three years, to 189,000 in 2021 (Chart 3.5). There are risks around that profile, however. Net migration in the year to 2018 Q2 was somewhat higher than the ONS’s principal projection, and that comparative strength could continue. But it is possible that these figures overstate the strength of net inward migration. Changes in employment by nationality from the Labour Force Survey (LFS) — an alternative source of data to those shown in Chart 3.5 — suggest that the number of EU citizens employed in the UK fell by 132,000 in the year to 2018 Q3.

In addition to population growth, labour supply growth also depends on structural changes in the number of people who want to work — those either in work or actively looking for a job.

The participation rate has risen slightly in recent years, as the proportion of people wanting to work has increased within certain demographic groups (the beige bars in Chart 3.6).² Improved health and longevity, as well as rises in the state pension age, have raised the participation rates of older workers. The proportion of women in or seeking work has also increased. Set against that, the rising average age of the UK population has pushed down the overall participation rate (the blue bars in Chart 3.6). Currently just over 10% of those aged over 65 participate in the labour market, relative to around 80% of those aged 16 to 64. These offsetting structural trends are expected to continue in coming years, such that the participation rate is projected to remain broadly stable.

The quantity of labour engaged in producing output will also depend on the equilibrium rate of unemployment. When unemployment falls below the equilibrium rate, wage and inflationary pressures will tend to build as companies need to pay more to recruit and retain staff.

The MPC judges that the equilibrium unemployment rate has fallen slightly in the past few years. In February 2018, it was estimated at around 4¼%. As discussed in previous Reports, the fall in the equilibrium unemployment rate is likely to have reflected structural factors such as changes to the tax and benefit system and an increased degree of educational attainment in the workforce.

The actual unemployment rate has fallen slightly over the past year (Chart 3.2), to a little below the MPC’s estimate of the equilibrium rate made in February 2018. The MPC judges that fall has reflected a cyclical rise in labour demand rather than a further fall in the equilibrium rate. The number of vacancies relative to the size of the workforce — a key indicator of labour demand — has risen to a historical high (Table 3.A). And the rate at which those already in employment are switching to new jobs — which will partly reflect the degree to which employers are competing to hire employees — has risen to close to its pre-crisis level (Chart 3.7). Stronger labour demand may also have reduced the job destruction rate since, in a tight labour market, workers become harder to replace (Chart 3.7).

The MPC judges that the equilibrium unemployment rate remains at 4¼%. That judgement is consistent with the recent strengthening in wage growth (Section 4), which has been slightly above the MPC’s projections in recent quarters and suggests that unemployment is close to, or below, its equilibrium.

Besides changes in the size of the labour force, potential labour supply can be affected by changes in the number of hours people want to work. Some part-time workers, for example, may prefer to find a full-time job, while others may want to reduce their hours. These preferences can be affected by current and expected future incomes, as well as demographic factors such as age.

In the decades prior to the crisis, average hours worked fell as incomes rose, since people could maintain their level of spending while working fewer hours. During the crisis, however, the share of part-time employment rose markedly and employees reported that, on average, they wanted to work more hours. Consistent with that, the proportion of part-time workers who reported that they could not find a full-time job rose (Chart 3.8). Since then, hours worked have risen towards their ‘desired’ level as the proportion of people working part-time has fallen (Section 3.1).

The MPC judges that desired hours worked per week will remain broadly stable over the forecast horizon, reflecting two offsetting structural factors. A rising average age of the population is likely to depress desired hours worked, since older people tend to want to work shorter hours. But set against that, desired hours worked by women are expected to rise further as more women work full-time.

Productivity

In addition to changes in potential labour supply, potential supply growth depends on gains in productivity. The level of UK productivity per hour is only just above its pre-crisis peak (Chart 3.9). By contrast, productivity in many other advanced economies is now some way above its level prior to the crisis.

The openness of the UK economy and the size of its financial sector mean that global developments, such as slower world trade growth and financial sector deleveraging, are likely to have been particularly important in driving the slowdown in UK productivity growth. Based on current data, the manufacturing sector — which tends to be particularly exposed to global growth — and the financial services sector can account for over half of the weakness in UK productivity growth since the crisis (Chart 3.10).³

In the financial services sector, the apparent slowdown in productivity growth partly reflects unusually high measured productivity growth prior to the crisis, driven by higher leverage and risk-taking within financial firms. Mismeasurement of financial services output may also have contributed to the measured slowdown, by overemphasising the effects of higher leverage before the crisis and the effects of deleveraging since then. While financial services productivity growth could pick up relative to the period following the crisis, the pace of growth seen in the 2000s is unlikely to return.

In the manufacturing sector, part of the slowdown in productivity growth is likely to have reflected the weakness in world trade growth since the crisis. World trade growth tends to boost productivity through increasing economies of scale, competition and exposure to new ideas. Although world trade growth picked up through 2017, it has since slowed (Section 1). All else equal, any ongoing weakness will weigh on the outlook for productivity growth in the manufacturing sector.

Another way to examine the productivity growth slowdown is to use a standard growth accounting framework to split productivity growth into the amount of capital available per hour worked — ‘capital deepening’ — and the efficiency with which both capital and labour are used to produce output — ‘total factor productivity’ (Table 3.C). Results from this approach suggest that slower growth in capital deepening can account for a significant part of the weakness in productivity growth since before the crisis. That has reflected weak investment over much of that period. The remainder of the weakness in productivity growth has been the result of slower growth in the efficiency with which inputs are used. This may partly reflect a misallocation of capital across both companies and sectors.

There are some signs that total factor productivity growth could pick up in coming years. Research and development (R&D) expenditure — a key driver of innovation and hence productivity growth — has increased as a share of GDP over the past decade to its highest level since the early 1990s (Chart 3.11). The extent to which that pickup in R&D spending will boost productivity growth is uncertain, however, and depends on a number of factors including the extent of complementary investment in tangibles — for example information technology — and intangibles — for example training and management. Furthermore, new discoveries tend to take time to implement and evidence suggests it can take between two and six years on average to boost productivity growth within firms.⁴

While the pickup in R&D spending could boost productivity growth in coming years, many of the factors that have weighed on productivity growth over the past decade are expected to persist. Business investment growth is expected to remain weak in the near term (Section 2), which will reduce the extent of capital deepening. Changes in trading arrangements as a result of Brexit are also likely to weigh on the outlook for productivity, even under the assumption of a smooth adjustment to those new arrangements.

Overall, the MPC judges that productivity growth is likely to be slightly weaker than previously projected. That downward revision has been made in light of the unexpected weakness in productivity growth over the past year. Output per hour — which was expected to grow by 1¼% in the year to 2018 Q4 under the MPC’s February 2018 projections — is now estimated to have been broadly flat (Chart 3.12). Productivity growth is nevertheless projected to pick up to around 1% per year by the end of the forecast period (Section 5).

• Notes

  ² The effects of demographics on participation and other aspects of the economy are discussed in Saunders, M (2018), ‘Some effects of demographic change on the UK economy Opens in a new window’.

  ³ For further details, see Tenreyro, S (2018), ‘The fall in productivity growth: causes and implications Opens in a new window’.

  ⁴ For further details, see Hall, B, Mairesse, J and Mohnen, P (2010), ‘Measuring the returns to R&D Opens in a new window’, Handbook of the Economics of Innovation.