Measuring the economic impact of Covid-19 in the UK with business - PowerPoint PPT Presentation

Measuring the economic impact of Covid-19 in the UK with business website data Juan Mateos-Garcia Starting at 11.30AM ESCoE COVID-19 ECONOMIC MEASUREMENT WEBINARS

Measuring the Economic Impact of Covid-19 with business website data Alex Bishop [@AlexJBishop] Juan Mateos-Garcia [@JMateosGarcia] ESCoE Covid-19 Economic Measurement Seminar 23 July 2020 nesta.org.uk @nesta_uk

Introduction Data and methods Findings Conclusion Summary We create a novel data pipeline to analyse sectoral and geographical ➢ exposure to Covid-19 in the UK. Our analysis shows that: Product search data tracks changes in consumer interest in products and services (and indirectly ฀ demand for industries) linked to Covid-19 Local economies with a larger proportion of the workforce in sectors negatively exposed to ฀ Covid-19 tend to have higher claimant count rates / faster growth in claimant count rates compared to pre-Covid 19 months. This link is intensified for locations with high shares of employment with low ■ diversification options away into less exposed sectors. Our semantic analysis of Covid-19 notices in business websites shows that they track the ฀ evaluation of the pandemic and reveals heterogeneity in company responses to it Our results suggest that novel data sources can help improve the evidence ➢ base about the economic impacts of Covid-19. Realising their value will require integration with other data sources (official and surveys) and ฀ innovation in how this information is disseminated. Click icons for interactive versions of the figures

Introduction

Introduction Data and methods Findings Conclusion Measuring the economic impact of Covid-19 Lockdown Supply Adaptation Impact Covid-19 Demand Scale Scope Change in process Social distancing Uncertainty Innovation What firms, sectors and places are most How are business reacting to this exposed to this shock? shock? Inform policies to mitigate Inform policies to adapt We need relevant, inclusive, timely and trusted indicators about these processes

Introduction Data and methods Findings Conclusion Measuring the economic impact of Covid-19

Introduction Data and methods Findings Conclusion State of play Data source Topic (see examples in the annex) Strengths and weaknesses Labour market Measure sectoral exposure to Covid-19 via scope to WFH in + Representativity ● different occupations + Comparability data - Lags - Explainability - Assumption of homogeneity Business panels Measure business adaptation and impacts via changes adopted / + Relevance ● expected and survival prospects + Comparability + Timeliness - Response biases - Small sample frames - Assumption of homogeneity “Big” data ● Measure exposure to Covid-19 through mentions in earning + Timeliness calls and their link with market value + Granularity ? Representativity Measure changes in demand with transaction data ● ? Comparability ? Reproducibility Measure sectoral exposure through changes in labour demand ● proxied via job ads Measure sectoral and geographical exposure through analysis of ● Covid-19 notices in websites

Data and methods

Introduction Data and methods Findings Conclusion Our setup Research questions: What firms, sectors and places are most exposed to the Covid-19 shock? ฀ What are their diversification options and what are they doing about them? ฀ We combine web sources with open and official sources to generate indicators of exposure to Covid-19 that are granular (at the firm, sector and geographical level), timely, comprehensive and comparable. 1.8 million websites in the UK collected in June 2020 (90% coverage of UK ➢ Business business websites according to Glass) websites (Glass) Obtained via web domain registries and enriched using machine learning. ➢ Contains business descriptions, predicted sectors, postcodes and text of ➢ Covid-19 notices Lacks direct information about exposure to Covid-19 and employment. ➢ Potential biases in coverage and noise in some information (eg registered ➢ addresses != trading addresses) Our coverage analysis (annex) shows strong correlation (r > 0.9) with ฀ sectoral and geographical distributions in CH and IDBR.

Introduction Data and methods Findings Conclusion Data pipeline We enrich Glass with other open and web sources to increase its relevance for our research questions. Gives us labels to Timely measure of Companies Generate local Nomis integrate web data exposure to Covid-19 estimates and House with official in terms of consumer validate with taxonomies interest measures of impact Enrich with SIC & location data Normalise and scale Measure sectoral Google Search Business Generate terms to and local Trends exposure to websites (Glass) query... Covid-19 Measure sectoral Analyse sector mix diversification in businesses options Granular descriptions of the many things that Measure Analyse Covid-19 businesses do adaptation to notices Covid-19 Work in progress! We look forward to your feedback

Introduction Data and methods Findings Conclusion Google Search Trends Blackfriars Scenery has unique experience in the area of live Established body of literature using Google Search data to events including awards ceremonies, theatre and performance staging. We are able to meet a wide range of needs by supplying proxy consumer / user interest in various subjects (not sets from our extensive stock of staging and flattage or by always successfully cf. Google Flu Trends) (see annex) producing settings tailored to your requirements. Our pipeline: 1. We aggregate company descriptions over sectors and extract salient terms (high frequency in division artistic_director arts_centre arts_council compared to corpus) focusing on 73 divisions with more theatre orchestra opera performing_arts comedy activity. dancers dance concerts arts jazz festival artists 2. We stem terms to remove duplicates 3. We query top 15 vs Google 4. We remove very low search frequency terms Observations Results will be dominated by most popular sub-sectors in ● Glass Noise in SIC codes is a limiting factor (“Other ● Manufacturing Activities ?) “I don't think that word means what you think it means” - ● see “dance” in search chart.

Introduction Data and methods Findings Conclusion Google Search Trends samples 01:Crop and animal production, hunting 11:Manufacture of beverages 25:Manufacture of fabricated metal and related service activities products, except machinery and equipment distillery brewery brewing cider ales gin breeding crops cattle seed growers farm metal_fabrication sheet_metal mild_steel beer drinks produce first great using steel_fabrication metalwork acres horses farmers plants animals local range new best quality one time precision_engineering powder_coating agricultural dogs estate land nursery years plating laser_cutting steelwork growing garden produce grown fabrication stainless_steel welding cnc aluminium metal steel wire tooling lead_times 62:Computer programming, consultancy 86:Human health activities 96:Other personal service activities and related activities chiropractic_clinic dental_care independent_funeral hair_salon barbers sql_server sharepoint custom_software orthodontics chiropractic funeral_directors hairdressing magento microsoft_dynamics cosmetic_dentistry dentures beauty_treatments salon funeral web_applications software_development dental_treatment chiropractors beauty_salon waxing hair_extensions open_source microsoft erp dental_practice dental_implants dry_cleaning hair_beauty stylists hair software_applications web_mobile osteopaths oral_health sports_injuries nails tanning tattoo grooming laundry mobile_apps salesforce ibm physiotherapy dentistry osteopathy physio web_development cloud_computing linux physiotherapists treatment_plan implementations oracle treatment_options

Findings

Introduction Data and methods Findings Conclusion Search trends (products) We measure changes in average consumer interest in 609 terms (~products and services) between February and April ( ฀ ) and February and June ( Δ ). Home consumption and production activities: gardening, baking, DIY Golf clubs before and afuer the lockdown Social consumption activities: theatre, travel, gym...

Introduction Data and methods Findings Conclusion Search trends (sectors) When we aggregate category search trends over SIC divisions we also find intuitive results. Decline in Real Estate, Accommodation and Food services and transportation (but note eg spikes in postal services and ฀ couriers) with recovery afuer the end of lockdown. Persistent decline in Arts, Entertainment and Recreation (excluding sports, which rebound) and Travelling ฀

Introduction Data and methods Findings Conclusion Sector changes We normalise search interest for categories in divisions in April / June by interest pre-Covid-19. The sector ranking suggests that: Industries involved in home production experience more search interest ฀ Industries involved in social consumption / construction experience less search interest ฀ Gardening and landscaping Retail of automotives services, manufacture of food, manufacture of wood, publishing, textiles, beverages. Activities of membership (eg religious) organisations Transport, accommodation, construction, personal health, creative arts...