National Statistical

Measuring the work of public services has probably never been as challenging as in recent months, with many services being reduced due to the pandemic while new services, such as track and trace, have been brought on stream to combat it. Here Jonathan Athow writes about how we have adjusted our previous estimates and how we expanding our statistics to include these new types of services. 

Government spending accounts for around a fifth of the economy, provides key services to millions of us and has been uniquely affected by the pandemic: no-where more so than in the provision of healthcare.  

For much of the economy we have a simple conceptual method for measuring the value of economic output. In most markets, we measure the value of output by adding up sales: the value of all new car sales, or the value of all the cups of coffee sold. We use this information along with the prices of new cars, cups of coffee and all other manner of goods and services to work out how much the quantity of cars or cups of coffee sold has changed. We account for changes in inputs used, weight these data together, and then estimate GDP: the amount of goods and services produced over a period.  

However, because many public service outputs are not paid for directly, meaning there is no market, this standard method does not work. There is simply no ‘market price for fixing broken legs’. Instead, our approach to estimating healthcare output is to directly count the quantity of activities undertaken: including attendances at A&E, elective surgeries, GP consultations and outpatient appointments. We combine these together using cost weights to estimate how much output has changed from one period to the next. This approach is regarded as best practice internationally, and one we use elsewhere for public sector production.  

The pandemic has had a significant and varied impact on NHS services. In some areas – such as critical care – activity has risen significantly as the number of Covid-19 hospitalisations has grown. However, the postponement or cancellation of many procedures to limit the spread of infection, as well as a fall in demand for some services, has held down healthcare output overall.  

This picture is replicated in other areas of activity – including the numbers of GP consultations, outpatient, elective and non-elective care activities – all of which fell sharply during the middle of 2020. Together with attendances at A&E, these activities account for almost three-quarters of the ‘weight’ of activity undertaken by the health service.  

By contrast, the service areas where the Covid impact has been largest carry the smallest weight: for instance, despite its high profile, critical care carries a weight of just 3.2 parts per hundred of NHS activity. As a consequence, healthcare output contributed 1.72 percentage points to the fall in GDP between Q2 2020 and Q2 2019.  

These indicators give a good idea of how healthcare output has evolved during 2020, but the complete picture of how the NHS has responded to the Covid-19 pandemic will only emerge gradually over time. In the short term, there is more uncertainty than usual about healthcare activity, and it will take some time before we have access to all the relevant data.  

To ensure that our data keep pace with developments on the ground, we have strengthened our relationship with the NHS. This has enabled us to increase the ‘data content’ of our first estimates of GDP: basing these early estimates on the most timely data on healthcare activity possible.   

For the first time tomorrow, our headline estimates of healthcare output will also include an explicit adjustment for the track and trace system. These will be early, indicative estimates which we will improve over time, and – given the small expenditure involved so far compared with total health spending – will have little impact on the overall level of health output. Because of the small size of the figures, they will also have little impact on the so-called ‘implied deflator’, which is the ratio of spending on healthcare to how many services are actually delivered. As in previous quarters, much of the recent growth in this deflator reflects the higher costs and fewer appointments in ‘regular’ healthcare, owing to the new infection control measures. 

 As we work on our Test and Trace estimates, we will also be laying the groundwork to capture the mass vaccination programme that began earlier this month.  

Amending and improving our approach to measuring the public sector through the pandemic has not been straightforward, but we are confident we are producing high quality estimates using the best internationally agreed rules.