Regional nowcasting in the UK

Timely and high-frequency regional macroeconomic indicators are essential for effective decision-making and supporting growth across the UK. By combining national and regional statistics produced by ONS with data produced by the devolved administrations in Scotland and Northern Ireland, our newly developed nowcasting model produces estimates of quarterly economic growth for all UK regions to the same approximate timetable as the release of UK GVA data. Our estimates are being used by policymakers at a sub-national level and are enabling new analysis and improved understanding of the pattern of historical regional growth.

There is a clear demand for better devolved, regional and local statistics in terms of quality, coverage, timeliness, frequency and granularity. Historically, the ONS produced regional output, specifically Gross Value Added (GVA), data at the annual frequency and with a significant release delay. Since September 2019, the ONS has released country and regional GDP data at the quarterly frequency around 6 months after the end of the quarter. Existing short-term indicators already exist for Northern Ireland and Wales, and Scotland has, for some time, produced its own quarterly GVA estimate.

Complementing this work, we have developed a nowcasting model that produces timely quarterly regional estimates of GVA. Our estimates are conditioned on both the historical annual data for the regions and the quarterly data for the UK. Since November 2018, our regional estimates have been released shortly after ONS’s first estimates of quarterly GVA for the UK. We have also produced historical quarterly estimates of regional GVA growth dating back to 1970.

The most recent iterations of this model now incorporate the ONS’s Regional Short-Term Indicators alongside Scottish GDP, and the Northern Ireland Composite Economic Index. We’ve also extended the model to incorporate a greater number of region-specific data in the form of a range of key economic predictors. We’ve also provided training for ONS staff to run these models.

We use mixed-frequency Bayesian Vector Autoregressive (BVAR) models, widely used to produce timelier and higher frequency estimates of a range of economic variables like economic growth. In broad terms, nowcasting methods using these models seek to exploit the mixed-frequency nature of the data and accommodate the differing publication timetables of those indicator variables chosen for their putative ability to explain the within-year or quarterly variable of interest, such as quarterly economic growth. A range of different econometric methods have been used to ‘nowcast’ macroeconomic variables. We use a state space VAR model. In order to make this model suitable for our application, we have explored different econometric innovations to meet our needs and, in particular, to impose the ‘cross-sectional’ constraint that the quarterly regional data add up to the observed UK data. We also explored the use of ‘machine learning’ to enable more efficient computation and deliver more accurate nowcasts of regional growth.

We have developed a nowcasting model that produces timely quarterly regional estimates of GVA to the same approximate timescale as UK quarterly GDP is released. The model is flexible in a number of ways, in particular its ability to incorporate additional predictors at a regional and national level. At its core it produces estimates of economic growth consistent with the UK and sub-national data produced by the ONS and devolved administrations.

In our academic papers we have explored the effect of a number of the model features on the accuracy of the point and density nowcasts produced by the model in a pseudo out-of-sample nowcasting exercise. A recurring feature of our results is the importance of having the data for the UK as a whole for a given quarter in producing more accurate regional estimates of that quarter, but we find that where the UK estimate is not available for a given quarter (as it has not been released yet), a range of model features including the cross-sectional restriction and additional regional predictors can prove valuable.

Our research has demonstrated that empirical macroeconomic methods are able to produce higher-frequency and more timely estimates of regional economic growth. This meets a key demand from stakeholders for more timely information on the performance of regional economies. Our methods also provide consistent historical regional growth data back to 1970 that enables current regional growth to be set into an appropriate historical context, helping inform policy and decision-making at the regional level.

Through the period from 2018 to 2021 we published our estimates in real-time on the ESCoE website, and the ONS are now using the model to generate estimates in real time for publication and use by the ONS and other users.

Our estimates have received media interest and have been cited by many sources in the private sector and by politicians in the UK Parliament and elsewhere.

Koop, G., McIntyre, S., Mitchell, J., and Poon, A. (2024) ‘Using stochastic hierarchical aggregation constraints to nowcast regional economic aggregates’ International Journal of Forecasting, Vol 40, Issue 2, Pages 626-640 https://doi.org/10.1016/j.ijforecast.2022.04.002

Koop, G., McIntyre, S., Mitchell, J., Poon, A. and Wu, P. (2024) ‘Incorporating short data into large Mixed frequency VARs for regional nowcasting’ Journal of the Royal Statistical Society: Series A, Vol 187, Issue 2, April 2024 Pages 477–495, https://doi.org/10.1093/jrsssa/qnad130

Koop, G., McIntyre, S., Mitchell, J., Poon, A and Wu, P. ‘Incorporating Short Data into Large Mixed-Frequency VARs for Regional Nowcasting’ Federal Reserve Bank of Cleveland Working Paper, WP 23-09, 5 August 2023, https://doi.org/10.26509/frbc-wp-202309

Koop, G., McIntyre, S., Mitchell, J., Poon, A. and Wu, P. ‘Regional Output Growth in the UK:Improving Estimates by Incorporating New Data Sources‘ ESCoE Conference on Economic Measurement 2023, Contributed Session N: Subnational measurement. King’s College London, 17-19 May 2023.

Koop, G., McIntyre, S., Mitchell, J., and Wu, P. ‘Timely Estimates of Household Income Growth and Income Inequality’ ESCoE Conference on Economic Measurement 2023, Contributed Session B: Households, inequality and unemployment. King’s College London, 17-19 May 2023.

Koop, G., McIntyre, S., Mitchell, J., Poon, A. and Wu, P. (2023) ‘Incorporating Short Data into Large Mixed-Frequency VARs for Regional Nowcasting’ ESCoE Discussion Paper Series, ESCoE DP 2023-07

Koop, G., McIntyre, S., Mitchell, J., Poon, A and Wu, P. ‘Regional Output Growth in the UK: Improving Estimates by Incorporating New Data Sources’ Economic Statistics Centre of Excellence: The Next Five Years, poster exhibition, 12 December 2022, One Birdcage Walk, London.

Koop, G., McIntyre, S., Wu, P., Mitchell, J. and Poon, A. ‘Sub-Regional Nowcasting of Economic Growth in the UK’ ESCoE Conference on Economic Measurement 2022, Poster Exhibition, University of Strathclyde, 26 May 2022

Koop, G., McIntyre, S., Mitchell, J. and Poon, A. (2022) ‘Using hierarchical aggregation constraints to nowcast regional economic aggregates’ ESCoE Discussion Paper Series, ESCoE DP 2022-04

“ONS publishes new, modelled estimates of growth across the UK’s national and regions using ESCoE’s ‘regional nowcasting’ approach” ESCoE News 8 October 2021

Rabaiotti, G. “Quick off the mark: How the ONS is speeding up estimates of regional GDP” ONS Blog, 7 October 2021

McIntyre, S. “UK regional growth – Q2 update” ESCoE Blog, 1 July 2021

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. (2021) “Nowcasting ‘True’ Monthly US GDP During the Pandemic” National Institute Economic Review, 256, pp44-70 doi.org/10.1017/nie.2021.8

Gary Koop, Stuart McIntyre, James Mitchell and Aubrey Poon (2021) “Regional Nowcasting in the UK” ESCoE Conference on Economic Measurement 2021, Poster Exhibition, Online, 11-13 May 2021. Poster Presentation

McIntyre, S., Mitchell, J. and Poon, A. ‘Nowcasting “True Monthly US GDP during the Pandemic” ESCoE Conference on Economic Measurement 2021, Contributed Session B: Nowcasting, Online, 11-13 May 2021 (presentation at 45:00)

McIntyre, S. “Regional impact of the pandemic – new estimates of regional output in 2020” ESCoE Blog, 18 February 2021

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. “Nowcasting ‘True’ Monthly US GDP During the Pandemic“Strathclyde Discussion Paper in Economics 21-1, Jan 2021

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. “Reconciled Estimates of Monthly GDP in the US” ESCoE Blog, 26 November 2020

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. (2020) “Reconciled Estimates of Monthly GDP in the US” ESCoE Discussion Paper Series, ESCoE DP 2020-16

McIntyre, S. “Regional growth in 2020 so far …” ESCoE Blog, 24 November 2020

Koop, G. “Reconciled Estimates of Monthly GDP in the US” presented at the Department of Economics, University of Salzburg, 20^th October 2020

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. “Improved Quarterly Estimates of Economic Activity in the Regions of the UK” ESCoE Conference on Economic Measurement 2020, Contributed Session N: Sub-national, Online, 16-18 Sep 2020

McIntyre, S. “Estimating UK regional growth in Q2 2020” ESCoE Blog, 28 August 2020

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. ‘Reconciled Estimates of Monthly GDP in the US‘ ESCoE Discussion Paper Series, ESCoE DP 2020-16, 26 Nov 2020

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. (2020). Reconciled Estimates and Nowcasts of Regional Output in the UK. National Institute Economic Review, 253, R44-R59. doi:10.1017/nie.2020.29

McIntyre, S. “Regional growth at the start of Covid-19: What happened across the UK in Q1 2020?” ESCoE Blog, 21 May 2020

Koop, G., McIntyre, S., Mitchell, J., Poon, A. (2020) “Regional output growth in the United Kingdom: More timely and higher frequency estimates from 1970“, Journal of Applied Econometrics Vol 35, Issue 2, Wiley doi.org/10.1002/jae.2748

McIntyre, S. “Reconciling regional growth data: what do the latest data tell us?” ESCoE Blog, 21 February 2020

Koop, G., McIntyre, S. and Mitchell, J. (2020) “UK regional nowcasting using a mixed frequency vector auto-regressive model with entropic tilting” Journal of the Royal Statistical Society, Statistics in Society Series A, Vol 183, Issue 1, Wiley doi.org/10.1111/rssa.12491

McIntyre, S. “Weakness in UK growth reflected in regional growth across the UK” ESCoE Blog, 21 November 2019

McIntyre, S. “UK regional growth in a time of economic uncertainty – estimates to 2019 Q2” ESCoE Blog, 16 August 2019

McIntyre, S. “Differences in economic growth across the UK remained substantial in the year to Q1 2019” ESCoE Blog, 20 May 2019

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. “Regional Output Growth in the United Kingdom: More Timely and Higher Frequency Estimates, 1970-2017” ESCoE Conference on Economic Measurement 2019, Contributed Session H: Nowcasting and forecasting, King’s College London, 8-10 May 2019

McIntyre, S. “Regional growth in 2018 – what happened?” ESCoE Blog, 19 February 2019

Koop, G., McIntyre, S., Mitchell, J., & Poon, A. ‘Regional Output Growth in the United Kingdom: More Timely and Higher Frequency Estimates, 1970-2017‘ ESCoE Discussion Paper Series, ESCoE DP 2018-14, 20 November 2018

McIntyre, S. “More timely and more frequent regional growth estimates” ESCoE Blog, 20 November 2018

Koop, G., McIntyre, S. and Mitchell, J. “UK Regional Nowcasting using a Mixed Frequency Vector Autoregressive Model” ESCoE Discussion Paper Series, ESCoE DP 2018-07, 12 Jun 2018

Koop, G., McIntyre, S. and Mitchell, J.”UK regional nowcasting using a mixed frequency vector autoregressive model” ESCoE Conference on Economic Measurement 2018, Contributed Session II: Aspects of the measurement of growth, Bank of England, 16-17 May 2018

For a full list of the quarterly updates visit here.