The UK’s capital services grew 1.3% in the fourth quarter of 2019 compared with a year earlier, the Office for National Statistics reported on 7 April 2020. Yet its account of the years since 2012 described capital services failing to keep pace with hours worked. There is no contradiction: the productive services available from assets can increase while the amount available per working hour falls. ONS calls that capital shallowing. The bulletin reports growth using natural-log changes, which can differ slightly from conventional percentage changes. [S1]

The distinction matters for interpreting the UK’s weak productivity figures. ONS estimated that multi-factor productivity in the market sector fell 0.3% in the fourth quarter of 2019 against the same quarter in 2018, using that logarithmic convention. This experimental measure tracks the part of output growth left after accounting for measured changes in capital services, hours and the composition of labour. It is not simply output divided by hours. The market sector also excludes general government and non-profit institutions serving households. [S1]

Software made the largest cumulative contribution to capital-services growth from the end of 2007 through the end of 2019: 3.5 percentage points, according to ONS. Capital services measure the flow of productive services from assets, rather than just the stock of assets. Short-lived assets such as software receive proportionately higher weights in this measure than they do in capital stocks, reflecting their more intensive use. The 3.5-point contribution therefore does not say that software raised output per hour by 3.5%, or establish a return on a software purchase. [S1]

For a business assessing investment, a useful question follows from the ONS framework: will productive services from its assets grow faster than working hours? That tests capital per hour, rather than assuming a purchase improves efficiency. Answering it requires evidence from the business, not the national aggregate. [S1]

ONS did not identify a single cause of the wider productivity slowdown. Its companion commentary set out competing explanations, including management practices, measurement problems and slower innovation. It said that a proposed single root cause must explain both the timing after the 2008 downturn and enough of the slowdown’s scale. Capital shallowing helps describe what happened to one input per hour; it does not, by itself, settle why productivity growth weakened. [S1][S2]

Although released in April 2020, these figures end in December 2019. ONS explicitly said the data were not affected by coronavirus. The agency warned that pandemic disruption could alter the relationship between working patterns and goods and services produced, requiring care with forthcoming releases. [S2]

Source documents

  1. ONS: Growth accounting: multi-factor productivity estimates, UK: October to December 2019https://www.ons.gov.uk/economy/economicoutputandproductivity/productivitymeasures/bulletins/growthaccountingmultifactorproductivityestimatesuk/octobertodecember2019
  2. ONS: Productivity economic commentary: October to December 2019https://www.ons.gov.uk/employmentandlabourmarket/peopleinwork/labourproductivity/articles/ukproductivityintroduction/octobertodecember2019