Measuring GDP in a digitalised economy

Measuring GDP in a digitalised economy

• 1.
  Jennifer Ribarsky Head ofSectoral National Accounts Section Statistics Directorate OECD Global Forum on Productivity 14 October 2016, London MEASURING GDP IN A DIGITALISED ECONOMY
• 2.
  Background 2 Increased prevalence of ‘new’transformative (digital) technologies But…. …. Declining productivity Market capitalisation of AirBnB (£ Billions) Trend labour productivity growth
• 3.
  Is it amis-measurement problem? Charlie Bean: “statistics have failed to keep pace with the impact of digital technology” The internet and the productivity slump The U.S. Underestimates Growth Charles Hulten: Valuing the Net and the wide range of applications… is challenging…. and their omission or undervaluation surely affects GDP.” Diane Coyle: The pace of change in OECD countries is making the existing statistical framework decreasingly appropriate for measuring the economy Why we’re measuring the digital economy in the wrong way Some optimists argue instead that the problem is one of measurement. Technological progress often raises productivity in ways that statistical agencies struggle to detect
• 4.
  • Based onan OECD working paper on Measuring GDP in a digitalised economy (N. Ahmad, P. Schreyer) – Issues and challenges of digitalisation • Aim: Take stock of current and best practices of OECD countries and key partners – 29 country responses (as of 1/10/2016) Survey of country practices
• 5.
  Digital intermediaries Digital intermediaries ForC2C Dwelling services, Business & Transport, Finance Conclusions: Conceptual framework sufficiently robust Estimation methods may need to be adjusted • A few countries capture digital rentals of dwelling services- mainly through tax records • More countries appear to capture Business and transport services- mainly from tax records and labour force surveys • Almost no country has estimates on financial intermediation services
• 6.
  Consumer durables andinvestment Household durables used in production Rise of the ‘occasionally’ self-employed: And use of durables in production Not a new phenomena and no impact on GDP Although MFP will be affected • But may be a need for more detailed guidance on delineation • A few countries make this distinction based on car registration, surveys, private market research, assumptions
• 7.
  Knowledge based capitaland globalisation ‘Investment’ outside of the SNA asset boundary and cross-border flows Many ‘intangible’ assets already in the SNA but many are not: Human capital, Knowledge in databases, Organisational capital , Brands And for those assets in the boundary, difficulties with cross-border transactions remain Not a new phenomena Considered in the 2008 SNA revision process but ruled out on practical grounds. Guidance developed in various Task Forces but further work may be needed as the scale of the problem remains unknown
• 8.
  Prices and volumes Asignificant challenge Customisation Outlet bias Quality change Not a new phenomena but challenges remain Price indices for software investment
• 9.
  • Majority treatas change in price, price differences in distribution margins occurring from buying products provided on-line versus in a store. • Only one country takes into account changes in quality (such as self-service checkouts at super markets) due to the increased participation of consumers in the production process. • Some countries are currently using or exploring new data sources, such as those that can be captured via web-scraping, to help deal with rapid quality changes and the emergence of new products as a consequence of digitalization. Prices
• 10.
  • For ownaccount software, many countries use some variation of an input method or use a services producer price index (SPPI) • For R&D, countries use an index of wage and salary earnings or in many cases taking into account all input costs (aka. Input method) • Input method does not capture quality or productivity changes (unless an explicit adjustment is made) Software and R&D prices
• 11.
  • That theconceptual framework is robust • But measurement in some areas may require improvement and new approaches for – The occasionally self-employed – International transactions in IPPs – And Prices • Whilst also recognising that complementary statistics to bridge gaps between consumer surplus and GDP may be necessary (satellite accounts) Conclusions:
• 12.
  Thank you