## 022818measuringdigitaleconomy ## Source details **Canonical URL:** [022818measuringdigitaleconomy](https://www.imf.org/-/media/files/publications/pp/2018/022818measuringdigitaleconomy.pdf) ## Other formats - [Markdown version](/-/media/files/publications/pp/2018/022818measuringdigitaleconomy.pdf.md) - [Structured JSON version](/-/media/files/publications/pp/2018/022818measuringdigitaleconomy.pdf.json) --- ### EXECUTIVE SUMMARY — Overview - Digitalization: new applications of information technology in business models and products transforming the economy and social interactions. - Digitalization acts as both an enabler and a disruptor of businesses. - Paper distinguishes: - “digital sector” (concrete perimeter: core activities of digitalization, ICT goods and services, online platforms, platform-enabled activities such as the sharing economy); - “digital economy” (broader: digitalization spread across all sectors). ### Definition and size of the digital sector - No generally agreed definition of “digital economy” or “digital sector”; classifications for Internet platforms and associated services are incomplete. - Evidence suggests the digital sector is still less than 10 percent of most economies by value added, income or employment. - Illustrative U.S. staff estimates for 2015 (output-based where noted): - ICT equipment, semiconductors and software: 2.8 percent of GDP - Telecommunication and Internet access services: 3.3 percent of GDP - Data processing, and other information services: 0.7 percent of GDP - Online platforms, including e-commerce platforms: 1.3 percent of GDP - Platform-enabled services (e.g., the “sharing economy”): 0.2 percent of GDP - Total (with incomplete adjustment for double counting of output): 8.3 percent of GDP - Conceptually not included in GDP, or missed for procedural reasons (illustrative U.S. figures): - Wikipedia and open source software: 0.2 percent of GDP - Free media from online platforms funded by advertising: 0.1 percent of GDP - “Do-it-yourself” fixed capital formation of online platforms: 0.3 percent of GDP - Output of MNEs attributed to tax havens: 0.4 percent of GDP - Total (with incomplete adjustment for double counting of output): 1.0 percent of GDP ### Measurement error in GDP growth and productivity statistics - Key sources of potential underestimation: - Insufficient adjustment for quality change in deflators for digital products. - Gaps in measuring the sharing economy and activities of online platforms. - Available research suggests effect on estimates of U.S. labor productivity growth is no more than 0.3 percentage points. - This estimated measurement effect is smaller than the post-2005 slowdown in productivity growth of 1–2 percentage points. - Symmetric effect on price statistics: slight overestimation of inflation if growth understated. - Symmetric inflation implication relevant for monetary policy assessment in economies with deflationary pressures during rapid digital transformation. ### Free digital services, GDP boundary, and welfare measurement - Proposals to include free digital services directly in GDP have been advanced. - Paper concludes changing GDP conceptual framework to directly include “free digital services” in consumption would not be warranted: - GDP measures market- and near-market production valued at market prices; suited to key policy questions. - Some free services represent quality improvements that could be captured via quality-adjusted deflators. - Research on expanding investment measures to include data collection may imply a modification of the GDP production boundary. - Indicators of welfare from free digital products should be developed outside the GDP boundary: - Productivity gains in households’ time use for nonmarket production may increase welfare not measured by consumption or GDP. - Debate about measuring household non-market production is increasingly pertinent. - International and national institutions should accelerate efforts to develop indicators of welfare growth from non-market production beyond GDP. ### Policy recommendations and data needs (summary) - Improve access by national statistics compilers to administrative data and to “Big Data.” - Administrative data: close cooperation of national government agencies. - Big Data: partnerships between private and public sectors, including international organizations. --- ### 11. The relevant weights for gauging sensitivity of estimates of GDP growth to mismeasurement of the digital sector — Digital-sector weights and digital transactions - Relevant weights for sensitivity are smaller than GDP shares because some output is intermediate consumption. - ICT services are particularly likely to be used as intermediate consumption. - Mismeasurement of digital-sector output used for intermediate consumption does not affect GDP (which equals final consumption + capital formation + net exports). - Alternative approach: define digital transactions (criteria include how the transaction is made, what is transacted, who is involved). - Expert groups’ working definition of digital transactions: products that are digitally ordered, digitally delivered, or platform-enabled. ### GDP versus welfare: conceptual framework - GDP measures production: three estimation approaches: - sum of value added of resident producers (adjusted for taxes and subsidies on products); - final expenditures (C + I + G + X–M); - income from production distributed to labor and capital or paid as taxes. - Free digital products raise concerns about neglected welfare gains; imputing them differs from actual money income and actual producer revenue. - Nominal GDP is not—and should not be confused with—a measure of welfare. - Consumer surplus from digital products is not recorded in GDP. - Online discrete choice experiment (U.S. Internet users, 2016) median annual consumer surplus estimates: - Search engines: $14,760 - E-mail: $6,139 - Digital maps: $2,693 - Online videos: $991 - E-commerce: $634 - Social media: $205 - Messaging: $135 - Music: $140 - TOTAL: 25,697 - Memo: Household disposable income per capita in U.S.: 43,469 - If U.S. household consumption were adjusted to include total consumer surplus of Internet users, its level would increase by about 30 percent. ### Real GDP, deflators, and measurement implications - Real GDP = nominal values deflated by a price index; characteristics of household consumption deflator are key. - Market prices yield weights that make growth of aggregate real consumption approximate welfare growth under usual assumptions. - Using hypothetical shadow prices to value free digital products is inconsistent with frameworks for welfare or productivity measurement. - Accuracy of deflation and quality-change adjustment central for measuring welfare and productivity. ### Production boundary and nonmarket production - SNA production boundary includes market and near-market production; volunteer services and households’ nonmarket production for own consumption are outside. - Example: nonmarket services of volunteers or for own consumption amounted to 36 percent of overall production (and 56 percent of GDP) in the United Kingdom in 2014. - Digitalization has shifted some market production outside GDP boundary (households acting as own travel agents, Skype replacing long-distance calls) and shifts in the opposite direction (gig economy, online shopping). - Free digital replacements change GDP composition, not necessarily level, as households may spend money saved on other items. - Capturing welfare from free digital replacements is primarily a price and volume index issue; many digital-product capabilities can be treated as quality changes in priced items. - Some volunteer-produced digital products and uses of digital products in household production raise production boundary questions. - Changing GDP definition would create more problems than it would solve; GDP suits policy questions (income, employment, monetary policy, potential government revenue, investment, productivity). - Replicability and objectivity favor current GDP definition; assigning monetary values to volunteer services or free online services is subjective. - Indicators “beyond GDP” can help understand welfare effects, distributional impacts, and changes in household time use. ### Proposals to modify measurement and outstanding research questions - Proposal: apply owner-occupied dwelling treatment to consumer durable goods (Byrne and Corrado, 2017): include services of consumer durables in household consumption adjusted for intensity of utilization—could significantly affect household consumption growth during rapid device uptake. - Impediments: practical problems, resources and source data, risk of obscuring business cycle developments (imputed services of durables would be smooth). - Proposals to change treatment of free media funded by advertising merit further research; current SNA treats free platforms funded by advertising as suppliers of advertising services. - Nakamura et al. (2017) proposal: expand output of platforms to include imputed media services consumed directly by viewers, value inferred from cost of production; outstanding questions include breadth beyond advertiser-funded media and whether expenses to attract users and collect data should be treated as investment in data. - Quantitative indication: Valuing free media based on advertising revenue implies increases in nominal GDP level in OECD countries in 2013 ranging from 0.4 percent of GDP in Greece to 1.3 percent of GDP in the United States. For growth, effects are close to zero or even negative. - Approaches quantifying benefits as perceived by consumers yield larger estimates but are not suitable for inclusion in market and near-market production measures. ### Globalization: geographic boundaries and firm relocations - GDP defined by production within a country’s economic territory; digitalization facilitates fragmentation and redomiciling of IP assets. - MNE relocations for tax reasons can affect reported location of production and BOP/IIP statistics. - Revisions to Ireland’s 2015 statistics highlighted potential effects: - Yearly GDP growth revised from 7.8 percent to 26.3 percent. - Goods exports revised up by €50 billion. - Net IIP revised from –€150 billion to –€532 billion. - Modified national concepts (example: Ireland’s modified GNI excluding factor income of redomiciled companies and depreciation of relocated assets) can provide insight if well-documented. - Apportionment technique (Guvenen et al., 2017) apportioning worldwide income of U.S.-headquartered MNEs based on labor inputs and sales implies: - increase of 1.7 percent in estimate of U.S. GDP level in 2012. - U.S. productivity growth increases by 0.1 percentage point in 1994–2004, and 0.25 percentage points in 2004–2008, with no change thereafter. - Upward revision to U.S. GDP could imply upward revision to net exports of goods and services. - Estimates include all industries; digital enterprises may account for less than half. ### Productivity measurement and the digital economy - Productivity measures ability of producers to transform inputs into output; only market producers are within scope. - Overstated deflators for ICT products identified as causing underestimation of real output and productivity. - Nonmarket output is outside standard productivity measures; digitalization improved households’ ability to produce nonmarket services, increasing welfare; measurement hampered by lack of time-use data. - Productivity = growth of market output (GDP) less growth of inputs (labor; for TFP, labor and capital). - Adjustments to real GDP to better capture digital output have slightly magnified effect on labor productivity growth because productivity base includes only market production. - For TFP, adjustments to output of fixed capital goods may have virtually no effect because capital is also an input. - Context: - Sharp slowdown in productivity in most AEs since the global financial crisis. - Some attribute slowdown to mismeasurement of the digital economy. - Growth of market output of digital sector probably underestimated due to underadjusted ICT prices, but ICT output is too small for much impact on aggregate productivity in most AEs. - Slowdown in productivity growth rate in AEs is more than 1 percentage point, and almost 2 percentage points in some cases. - Underestimation of labor productivity growth due to underadjustment for quality change in ICT goods and services is generally under 0.3 percentage points—or less if mismeasurement before slowdown is considered. - Effects on TFP likely smaller. - Additional quantitative context: - Output of ICT goods and services is about 6 percent of U.S. GDP in Table 1. - Almost half of this is software, whose growth rate appears underestimated only slightly, perhaps by 3 percentage points or less. - Quality improvement in computers and semiconductors seems underestimated, perhaps by as much as 10 percentage points, but their weight in market output covered by productivity calculations is just 0.5 percent. --- ### 35. PRICE INDEX COMPILATION CHALLENGES AND STATE OF PLAY — Key challenges - Four challenges for price indexes deflating digital GDP components and measuring inflation: - Capturing fast-changing quality of digital goods and services. - Handling introductions of new products. - Measuring e-commerce. - Measuring the sharing economy. ### A. Quality Adjustment and Price Indexes - Quality improvements represent increases in real output; price indexes must adjust prices for model differences or use techniques that avoid cross-model comparisons. - Matched models technique excludes non-matching observations; assumes quality change reflected in new-versus-old price differential when models sold side-by-side. - iPhone illustration (staff calculations, April 2017 prices and Apple’s 2016 Annual Report): - iPhone 6s 32GB transactions price: 2016 = 550; 2017 = 450. - iPhone 7 32GB transactions price: 2017 = 650. - Quality adjustment (iPhone 7 vs iPhone 6s): 200. - Quality-adjusted iPhone 7 price: 450. - Quality-adjusted Price Index for iPhones: 2016 = 100; 2017 = 81.8. - iPhone Nominal Output, Western hemisphere ($ US billions): 2016 = 54.9; 2017 = 60.5. - Nominal Value Index: 2016 = 100; 2017 = 110.2. - iPhone Sales, in constant prices of 2016: 2016 = 54.9; 2017 = 73.9. - Volume index: 2016 = 100; 2017 = 134.7. - Volume index with no quality adjustment (90.6 million in 2017/ 85.1 million in 2016): 2016 = 100; 2017 = 93.2. - Result: In 2017, iPhone nominal output growth = 10.2 percent while real output growth = 34.7 percent after deflating by quality-adjusted price index 81.8. - New goods: truly novel goods may be incomparable to existing goods; welfare gains from such appearances concentrated in short rapid-uptake periods. - Delays in updating index baskets can cause overstatement of price change; international guidelines allow updating of basket structure every 5 years. - IMF-OECD 2017 survey (43 responses, 33 from OECD): just three countries make no adjustments for quality changes in ICT and high-tech products. - Sample rotation practices can overstate price change for models entering during rotations (Moulton, Moses and Lafleur, 1998 found 2 percentage point overstatement for televisions). - Magnitude and cross-country variation: - Combined effects of underestimation of quality change in digital products on consumer inflation in an average OECD economy may be around 0.3 percentage points (see Box 2). - Ahmad et al. (2017) adjusted growth rates over 2010–15: adjusted indexes for Germany and France fall more than 4 percent per year; Belgium, Italy, and the United Kingdom rise or are flat; remaining countries fall by 2 to 3 percent per year. - Abdirahman et al. (2017): 7-percentage point overstatement of price growth rate in telecom services in the United Kingdom in 2010–2015. - Internet access and online content: faster connection speeds and growth in online content are quality improvements; cybersecurity issues can be quality declines. - Recommendations for NSOs: - Quality-adjust prices of key digital products drawing on other countries’ experience. - Consider innovations in data sources, collection, and index calculation to incorporate new varieties and suppliers quickly. - Maintain up-to-date samples of varieties and suppliers. - Foster close collaboration between price statisticians and national accounts compilers. - Follow Voorburg Group and Ottawa Group recommendations. ### Box 2 — Quantified effects on inflation and cost of living (recalibrations summary) - Reinsdorf and Schreyer (2017) calibration assumptions using OECD PPP program weights: - Assume ICT equipment and telecommunication services indexes overstate price growth by 5 percentage points due to underestimation of quality improvements. - Assume other goods incorporating digital technology (televisions, automobiles) overstate price growth by 2 percentage points. - Based on 2015 consumption basket weights, household consumption deflator would overestimate inflation by 0.28 percentage points from these effects. - For categories with free or inexpensive digital replacements, assume overstatement by 5 percentage points; weakly affected categories assume 2 percentage points. - Effect on household consumption deflator: 0.18 percentage points with 2005 weights, falling to 0.11 percentage points with 2015 weights. - Marginal benefits of expanded variety and better selection could be 0.3 percentage points overstated at product level, implying 0.06 percentage point impact on household consumption deflator. - Combined 2015 bounds for three effects: upper bound of 0.45 percentage points for combined overstatement of rate of change in cost of living. - Context: productivity slowdown reduced growth by more than 1 percentage point in the AEs. ### B. Coverage of E-Commerce and the Sharing Economy - E-commerce adoption: - U.K. ONS: non-gasoline retailers in the United Kingdom in 2017 made more than 16 percent of their sales online. - Eurostat (EU, 2016): 66 percent of household Internet users made online purchases; businesses with 10+ employees reported sales over the Internet to consumers = 2.7 percent of turnover and all other e-commerce sales = 15.6 percent of turnover. - Price implications: - Cavallo (2017): Amazon prices on average 5 percent lower than offline stores. - Adobe Digital Price Index vs U.S. CPI (2014–2017): DPIs fall by 1 percent per year, on average, relative to corresponding CPIs (Goolsbee and Klenow, 2018). - Billion Prices Project data generally show similar rates of change in online and offline prices (Cavallo and Rigobon, 2016). - IMF-OECD survey (43 countries) coverage of e-commerce in price indexes: - Domestic e-commerce included in CPI: 30 (69.8) - Domestic e-commerce included in PPI: 13 (35.1) - Cross-border e-commerce included in CPI: 12 (27.9) - Cross-border e-commerce included in PPI: 7 (18.9) - Regional breakdowns provided in survey results. - Finding: about 70 percent of respondents include e-commerce prices in CPI; PPI coverage limited. Gaps for products frequently purchased online (clothing and footwear); lags in reflecting fast-changing purchasing patterns. - Priority: adequate coverage of e-commerce in price samples and timely sample updating. --- ### 51. Sharing economy suppliers and CPI inclusion - Sharing economy suppliers should be included in the CPI with weights reflecting their importance in consumers’ spending patterns. - Where significant substitution to lower-priced sharing-economy replacements has occurred, estimating effect on cost of living is important for welfare measurement and CPI accuracy. - If replacement is a close substitute (example: Uber vs taxis), CPI may reasonably reflect change in average price paid. - Progress: only three IMF-OECD survey respondents (Australia, Germany and the United States) include sharing-economy prices in their CPI; none include them in PPI. - Measurement error depends partly on relative size of sharing economy, which still tends to be small; many respondents regarded sharing economy as relatively unimportant. ### National accounts compilation challenges — A. E-Commerce and Free Products from Online Platforms - 2016 OECD survey (29 responses) and 2017 IMF extension (11 responses) showed slow progress in estimating the digital sector; many countries do not prioritize measurement due to lack of resources and source data. - Malaysia developing an ICT satellite account including online platforms. - Only a third of surveyed countries collect data on online purchases; just five collect separate data on cross-border e-commerce transactions. - Ghana, India, Jamaica and Malaysia reported including e-commerce data in national accounts compilation. - Proposals to impute viewers’ consumption of free online media funded by advertising are under discussion; just one country developed experimental estimates. - Only eight respondents agreed free products funded by advertising should be included in household consumption or a new final consumption category. - No country has data to impute production by volunteers of free online content/software nor information on who consumes them. - Lack of coverage of open source software is a statistical concern because it affects measurement of TFP and commercial software production. - GDP compilers often base software production estimates on input costs, including software coder earnings; freemium business models complicate attribution. ### National accounts compilation challenges — B. The Sharing Economy - Sharing-economy suppliers may be informal; national accounts procedures for informal rental and labor reflect pre-sharing-economy conditions. - Household surveys often under-report sharing-economy work unless explicit questions included. - Progress on bringing platform-enabled rentals and labor services into tax, regulatory and reporting regimes should improve GDP measurement. - Only six OECD-survey countries capture property rentals through digital intermediaries; in IMF survey only India reported capturing these rentals. - Information sources: tax data or direct collection from digital intermediaries. - Relative scale: worldwide host revenue estimate (press reports of Airbnb 2017 revenue forecast $2.8 billion) suggests host revenue perhaps $30 billion worldwide. - GDP includes imputation for services of owner-occupied dwellings capturing part of peer-to-peer rentals. - For labor services, in about half the countries where legal, GDP compilers include sharing-economy transportation services in self-employment income estimates based on tax data (8 countries) or labor force surveys (7 countries). - Intermediation fees retained by platforms are challenging to measure; only five countries currently capture these fees. - Often intermediation services supplied by foreign platforms do not report cross-border transactions; available data may lack detail on intermediation fee components. ### National accounts compilation challenges — C. Lags and Data Gap Concerns - International guidelines recommend updating benchmark year at least every five years; rapid online platform growth may make five-year intervals underestimate sector size. - Composition of available deflators often lags composition of aggregates they deflate; mismatch can cause measurement error. - Data users need more extensive, granular statistics on scale and structure of digital activity; alternative classifications may be required (example: employment in e-commerce retailing may include establishments classified in warehousing). - Ahmad and Ribarsky (2017) discuss work on guidelines for a digital economy satellite account. ### Compilation challenges and state of play in external sector statistics — A. Digital Trade - Digital trade measurement initiatives underway (cross-border e-commerce project; collaboration on guidelines). - Preliminary measurement framework for digital trade considers multiple dimensions and transaction types. - Digitally-delivered products include software, media and cross-border data flows (e.g., advertisements). - 2014 ICT services (and potentially ICT-enabled services) shares: - United States: 54 percent of services exports and 48 percent of services imports. - EU: 56 percent of services exports to non-EU countries, and 52 percent of services imports from non-EU countries. - Emerging market and developing economies in 2014: 33 percent of services exports and 27 percent of services imports. - Digital ordering, payments, and delivery foster small transactions below customs thresholds; more than half of OECD countries include estimates of below-threshold transactions in IMTS; a minority of non-OECD countries do so. - Largest estimate of below-threshold trade share among IMF respondents: about 15 percent. - U.S. estimates: 2.2 percent of goods imports and 0.8 percent of goods exports in 2015 attributed to below-threshold adjustments. - Russia example: share of e-commerce in imports grew from 2.5 percent in 2016 to 4.2 percent in Q1 2017. - Few countries have separate estimates of cross-border e-commerce. - Digital downloads covered in most OECD countries and almost half of non-OECD countries, though not always separately identifiable. - Eleven countries indicated potential capture of household consumption of cross-border services through credit card data. - Many respondents viewed trade in digitally-delivered services as under-reported, particularly imports. - Intermediation services of non-resident platforms rarely measured: 15 percent of OECD countries and 6 percent of non-OECD countries can identify payments to nonresident digital intermediaries. - Foreign-owned digital intermediaries hard to identify on business registers in three-quarters of OECD countries and nearly 90 percent of non-OECD countries. - Cross-border unpriced data flows not measured; most respondents opposed imputing values of cross-border data flows in balance of payments for conceptual and practical reasons. ### Compilation challenges and state of play in external sector statistics — B. Digital Payments and Cross-Border Remittances - Standard BOP methods may omit some payments using digital technologies; implications for remittances. - Table 5: Global Totals of Remittance Receipts and Payments (Billions of U.S. Dollars) - Credit: 2009 397.3; 2010 424.2; 2011 472.6; 2012 496.1; 2013 525.4; 2014 555.9; 2015 550.8 - Debit: 2009 303.3; 2010 302.9; 2011 340.6; 2012 356.3; 2013 390.4; 2014 400.5; 2015 386.0 - Discrepancy: 2009 94.0; 2010 121.3; 2011 132.0; 2012 139.8; 2013 135.0; 2014 155.4; 2015 164.8 - Mobile money has significant share of cross-border transfers, especially between neighboring countries; can offer transaction cost savings exceeding 50 percent (GSMA, 2016a). - New digital channels (mobile money, online platforms, Bitcoin) may exacerbate source-data gaps for remittances. - Common data sources: bank and money transfer operator reports, household surveys. - World Bank central bank survey (2008–2009): 43 percent of remittance-receiving countries collect information on remittances via informal channels. - In countries with widespread mobile money (Uganda, Kenya, Philippines), no data collected from telecoms on cross-border remittances via their networks; no adjustments made. - Data accuracy could improve as digital transfers replace informal mechanisms and digital transfer platforms enter regulatory/reporting regimes. - Measuring mobile money remittances is an area for research; surveying telecommunications companies is a proposed solution. --- ### 82. DIGITALIZATION IN THE FINANCIAL SECTOR — Fintech components and measurement implications #### A. Marketplace Lending Platforms - Marketplace lending platforms: facilitate peer-to-peer (P2P) credit transactions or lend own funds. - P2P platforms primarily match borrowers with individual investors; may allow banks/institutional investors to invest in loans. - Some marketplace lenders securitize loans via bank arrangements (example: SoFi and Prosper). - Geography: large lending volumes in China, the United States, and the United Kingdom. - Benefits: improved access to credit and lower cost for SMEs and households; use non-traditional credit scoring. - Measurement/classification issues: - Less regulated than conventional intermediaries, may not be subject to reporting requirements, delaying inclusion in MFS data and requiring direct surveys. - MFS compilation guidelines classify P2P platforms as financial auxiliaries when they do not assume credit risk and rely on fees. - Loans made by households with assistance of these auxiliaries currently not included in MFS; household P2P lending may need supplementary reporting. - Marketplace lenders that lend own funds should be included as financial intermediaries, but data may be unavailable. #### B. E-Money - Definition: monetary value, claim on issuer, electronically stored on card/device/server, used for payments to third parties; includes mobile money, widely-accepted pre-paid cards, web-based products. - Exclusions: digital currencies; credit/debit cards; mobile phone payment apps linked to bank accounts; store-specific pre-paid cards. - Role: low share in money supply and payment transactions but important for financial inclusion. - Mobile money presence: in more than 90 countries as of 2015; in 21 African countries mobile money accounts approached or exceeded number of bank accounts. - Measurement/data implications: - MFS accuracy largely unaffected because issuers mirror outstanding e-money in escrow accounts at regulated financial institutions. - Mobile money operators offering non-payment services pose no measurement problems if partnered with regulated financial providers. - Mobile money creates data opportunities to improve household statistics using provider data. - Selected data points (Share of E-money Transactions in Noncash Payments in 2015) examples: - Austria 0.01; Germany 0.00; Greece 0.07; Italy 0.21; Singapore 0.24; Nigeria 0.94 - Mobile Money Share of Accounts (percent of total number of accounts at mobile phone companies and commercial banks, 2015 or most recent): - Afghanistan 6.04; Bangladesh 30.60; Benin 50.95; Botswana 51.19; Burkina Faso 47.76; Cambodia 14.33; Cameroon 79.62; Cote d'Ivoire 78.10; Egypt 12.33; Ghana 53.31; Kenya 47.34; Madagascar 65.21; Mali 63.61; Mozambique 42.42; Namibia 33.13; Nigeria 13.75; Pakistan 25.40; Philippines 19.61; Rwanda 84.84; Senegal 65.00; Tanzania 87.40; Uganda 81.89; Vietnam 6.24; Zambia 65.29; Zimbabwe 91.51 #### C. Digital Currencies - Definition: means of payment existing only electronically; cryptocurrencies such as Bitcoin are best-known. - Over 1,300 digital currencies exist. - Early 2017 estimates: between 2.9 and 5.8 million unique active users of digital currency wallets (Hileman and Rauchs, 2017). - Potential implications: could affect measurement of liquidity if widely accepted as medium of exchange; central banks investigating issuance of digital currencies. - Current classification: do not qualify as money in MFS framework; classified as nonfinancial assets. - Bitcoin examples: - March 2017 Bitcoin market capitalization: US$25 billion, compared with US$1.5 trillion of U.S. dollar currency in circulation. - February 2018 Bitcoin capitalization: almost $150 billion. - If digital currencies become widely accepted, consideration would be given to inclusion in broader liquidity measures. --- ### Policy implications and consolidated recommendations - High-level findings: - Slow productivity growth since the financial crisis is genuine and requires policy responses, not only measurement fixes. - Underestimation of digital sector output could subtract 0.3 percentage points from productivity growth rate, versus a 1- to 2-percentage point slowdown. - Symmetric effect on price statistics can yield slight overestimation of inflation due to under-adjusted quality change and lags in new products/suppliers inclusion. - Digitalization has exacerbated compilation weaknesses and created new data needs across macroeconomic statistics; potential need to consider adjustments to GDP production boundary and alternative treatments in national accounts and BOP. - Institutional roles: - Fund positioned to promote international cooperation and research on definitions, classifications and measurement techniques under Article VIII, Section 5. - International organizations should update classification systems and develop measurement guidelines. - Fund should be ready to respond to member country requests for technical assistance on digital economy measurement. - General recommendations: a) Improve coverage of digital platforms and services linked to digital platforms in main classification systems; and b) Develop aggregate classifications covering the digital sector, digital products and digital transactions. - Resources and data access: - National statistical offices require additional resources and better access to source data to implement improvements and disseminate indicators, including welfare and nonmarket production linked to digitalization. - Declining survey response rates and emerging new sectors make expanding use of administrative data and “Big Data” essential. - Governments should share data needed for statistics; national and international organizations should facilitate access to Big Data via partnerships with global firms. - Recommendations: a) Endow national statistical offices with sufficient resources to measure digital products and develop welfare indicators; and b) Ease access by national statistical compilers to administrative data and Big Data; promote data sharing and public–private partnerships. - Recommendations for price statistics: - Focus on improving quality-adjustment procedures for ICT goods and services; timely inclusion of new digital product varieties and suppliers; timely inclusion of new digital products in basket and weighting structures. - Recommendations: a) Statistical agencies should quality-adjust a selective list of products drawing on other countries’ work; b) International organizations should develop fit-for-use compilation approaches for compilers with resource constraints; and c) Statistical agencies should consider innovations in data sourcing, collection, and processing to include new digital products promptly. - Recommendations for national accounts: - Key challenges: deflation and complete coverage of digital platforms and platform-enabled activity. - Recommendation: a) National accounts compilers and price statisticians should collaborate to align composition of deflators for digital products with composition of aggregates being deflated, and where possible use datasets with prices and quantities to simultaneously calculate deflators and nominal values. - Recommendations for external sector statistics: - Challenges: growth of small transactions and cross-border services/payments using digital platforms. - Recommendations: a) Update assumptions concerning small transactions facilitated by digital ordering and delivery; b) Enhance collection of information on cross-border services provided by or through online platforms; and c) Develop methods for estimating international payments via new digital channels, such as mobile money remittances. - Recommendations for monetary and financial statistics: - Fintech-generated liquidity and credit could become important in the future. - Recommendations: a) Add marketplace lending platforms that lend own funds to credit statistics, and report supplementary data on peer-to-peer lending; and b) Investigate methods for compiling statistics on digital currencies. *Source: 022818measuringdigitaleconomy (selected sections and executive summary).* ### EXECUTIVE SUMMARY ### EXECUTIVE SUMMARY ### Overview - Digitalization encompasses a wide range of new applications of information technology in business models and products that are transforming the economy and social interactions. - Digitalization is both an enabler and a disruptor of businesses. - The paper distinguishes between the “digital sector” and the increasingly digitalized modern economy (the “digital economy”), and focuses on measurement of the digital sector, which covers: - the core activities of digitalization, - ICT goods and services, - online platforms, - platform-enabled activities such as the sharing economy. ### Definition and size of the digital sector - There is no generally agreed definition of the “digital economy” or “digital sector,” and industry/product classifications for Internet platforms and associated services are incomplete. - The term “digital sector” refers to a concrete perimeter of economic activities; the term “digital economy” is often used more broadly to indicate that digitalization has spread across all sectors. - Available evidence suggests that the digital sector is still less than 10 percent of most economies if measured by value added, income or employment. - Illustrative U.S. staff estimates for 2015 (output-based where noted) show: - ICT equipment, semiconductors and software: 2.8 percent of GDP - Telecommunication and Internet access services: 3.3 percent of GDP - Data processing, and other information services: 0.7 percent of GDP - Online platforms, including e-commerce platforms: 1.3 percent of GDP - Platform-enabled services (e.g., the “sharing economy”): 0.2 percent of GDP - Total (with incomplete adjustment for double counting of output): 8.3 percent of GDP - Conceptually not included in GDP, or missed for procedural reasons (illustrative U.S. figures): - Wikipedia and open source software: 0.2 percent of GDP - Free media from online platforms funded by advertising: 0.1 percent of GDP - “Do-it-yourself” fixed capital formation of online platforms: 0.3 percent of GDP - Output of MNEs attributed to tax havens: 0.4 percent of GDP - Total (with incomplete adjustment for double counting of output): 1.0 percent of GDP ### Measurement error in GDP growth and productivity statistics - Key sources of potential underestimation: - Insufficient adjustment for quality change in deflators for digital products. - Gaps in measuring the sharing economy and activities of online platforms. - Available research suggests the effect of under-measurement of the digital sector on estimates of U.S. labor productivity growth is no more than 0.3 percentage points. - This estimated measurement effect is smaller than the post-2005 slowdown in the growth in productivity of 1–2 percentage points. - There is a largely symmetric effect on price statistics, yielding a slight overestimation of inflation: - If growth has been understated due to insufficient downward adjustment of price indexes in the presence of high quality increases in digital products and services, inflation must generally have been overstated by a roughly similar amount. - The symmetric inflation implication is particularly relevant for assessing the monetary policy stance in economies that have suffered deflationary pressures while experiencing rapid digital transformation. ### Free digital services, GDP boundary, and welfare measurement - Proposals to directly include free digital services (self-produced, volunteer-produced, or produced by platforms that sell advertising and collect users’ data) in GDP have been advanced. - The paper concludes a change in the conceptual framework of GDP to directly include “free digital services” in consumption would not be warranted: - GDP is a measure of market- and near-market production valued at market prices and is suited to key policy questions. - Some free services enabled by digital products represent quality improvements that could be captured in real consumption by quality-adjusting the deflator. - Research on expanding the measure of investment to include collection of data may imply a modification of the GDP production boundary. - Indicators of welfare from free digital products should be developed in the context of measurement of nonmarket production outside the GDP boundary: - Productivity gains in households’ time use for nonmarket production may be increasing welfare in ways not measured by consumption or GDP. - The debate about measuring household non-market production is now even more pertinent. - International and national institutions need to accelerate efforts to develop indicators of welfare growth from non-market production beyond the boundary of GDP. ### Policy recommendations and data needs - Recommendations include improving access by national statistics compilers specifically to administrative data and generally to “Big Data.” - For administrative data, this entails close cooperation of national government agencies. - For Big Data, cooperation should extend to partnerships between private and public sectors, including international organizations. *Source: 022818measuringdigitaleconomy - EXECUTIVE SUMMARY* ### 11. The relevant weights for gauging sensitivity of estimates of GDP growth to ### 11. The relevant weights for gauging sensitivity of estimates of GDP growth to mismeasurement of the digital sector ### Digital-sector weights and digital transactions - Relevant weights for gauging sensitivity of estimates of GDP growth to mismeasurement of the digital sector are smaller than the GDP shares in Table 1 because some of the output is used for intermediate consumption. - ICT services are particularly likely to be used for intermediate consumption. - Mismeasurement of the output of the digital sector used for intermediate consumption by business would not matter for estimation of GDP, which equals final consumption, plus capital formation, plus net exports. - Defining digital transactions is an alternative to defining the digital sector. Possible criteria for distinguishing digital transactions include: - how the transaction is made (digitally ordered, enabled or delivered), - what is transacted (goods, services or data), - who is involved (consumer, business or government). - The expert groups’ current working definition of digital transactions includes products that are digitally ordered, digitally delivered, or platform-enabled. ### GDP versus welfare: conceptual framework - GDP is a measure of production, specifically market and near-market production valued at market prices. Three approaches to estimate GDP: - aggregating the value added of all resident producers (and adjusting for taxes and subsidies on products), - adding final expenditures on household consumption, capital formation, government consumption, and net exports (C + I + G + X–M), - adding the income from production distributed to the suppliers of labor and capital or paid as taxes. - Free digital products raise concerns about neglected welfare gains. Adding free digital products to output and consumption would require imputing transactions that raise gross income of the consumer and producer, but imputed income differs in important ways from actual money income and imputed producer revenue differs from actual revenue. - Nominal GDP is not—and should not be confused with—a measure of welfare. - Total welfare from digital products includes consumer surplus, which is not recorded in GDP. - Online discrete choice experiments (Brynjolfsson et al., 2017) found large median amounts U.S. Internet users would accept to forego digital products. Median annual consumer surplus estimates (Internet users in the United States in 2016) are: - Search engines: $14,760 - E-mail: $6,139 - Digital maps: $2,693 - Online videos: $991 - E-commerce: $634 - Social media: $205 - Messaging: $135 - Music: $140 - TOTAL: 25,697 - Memo: Household disposable income per capita in U.S.: 43,469 - If household consumption in U.S. GDP were adjusted to include the total consumer surplus of Internet users, its level would increase by about 30 percent. ### Real GDP, deflators, and measurement implications - “Real GDP” statistics are calculated by deflating nominal values by a price index; the characteristics of the deflator for household consumption are key. - Market prices used in national accounts yield weights that make growth of aggregate real consumption a measure of overall welfare growth under usual assumptions. - Proposals to use hypothetical shadow prices to value free digital products are inconsistent with the underlying conceptual frameworks for measuring welfare growth or productivity growth. - Accuracy of deflation and adjustment of prices for quality change are central for measuring welfare and productivity, since productivity gains often lower prices. ### Production boundary and nonmarket production - GDP production boundary (SNA) includes market and near-market production; volunteer services and households’ nonmarket production for own consumption are outside the boundary. - Example: Nonmarket services of volunteers or for own consumption amounted to 36 percent of overall production (and 56 percent of GDP) in the United Kingdom in 2014. - Digitalization has enabled shifts of market production to outside the GDP boundary (e.g., households acting as their own travel agent, Skype replacing long distance calls), but shifts also go in the opposite direction (e.g., gig economy reducing transaction costs, online shopping shifting tasks to market producers). - Shifts across the production boundary caused by technology, social change, or economic development occurred before the digital age; discussions must consider broader household non-market production context. - Free digital replacements change the composition of GDP, not necessarily its level, because households are likely to spend money saved on other items. - Capturing welfare from free digital replacements is primarily a price and volume index problem rather than a production boundary problem; many digital-product capabilities that eliminate household expense can be treated as quality changes in priced items (e.g., smartphone camera as quality attribute). - Some digital products from volunteers and uses of digital products as inputs in household production raise production boundary questions. - Changing the definition of GDP fundamentally would create more problems than it would solve: current GDP suits policy questions involving income, employment, monetary policy, potential government revenue, investment and productivity; nonmarket household production does not generate spendable income easily taxed or used to fund investment. - Replicability and objectivity favor the current GDP definition since it can be estimated from observable transactions and market prices; assigning monetary values to volunteer services or free online services involves subjectivity. - Indicators “beyond GDP” can help understand welfare effects of digitalization, including impacts on different population segments and changes in household time use. ### Proposals to modify measurement and outstanding research questions - Proposal: apply current treatment of owner-occupied dwellings to consumer durable goods, including digital devices; Byrne and Corrado (2017) would include services of consumer durables in household consumption, adjusted for intensity of utilization—this could significantly affect household consumption growth during rapid device uptake. - Main impediments: practical problems, resources and source data, and possible obscuring of business cycle developments (imputed services of durables would be smooth). - Proposals to change treatment of free media funded by advertising merit further research. Current SNA treatment: free platforms whose revenue comes from advertising are suppliers of advertising services; viewers pay as part of advertised-product prices. - Nakamura et al. (2017) would expand output of these platforms to include imputed media services consumed directly by viewers, with value inferred from cost of production; outstanding questions remain: - Free services to attract platform users is broader than advertiser-funded media; broader implications need understanding. - Whether expenses to attract platform users and collect data should be treated as investment in data — data as a product or asset should be part of the research agenda. - Quantitative indication: Valuing free media based on advertising revenue implies increases in level of nominal GDP in OECD countries in 2013 ranging from 0.4 percent of GDP in Greece to 1.3 percent of GDP in the United States. For growth, effects are close to zero or even negative. - Approaches that quantify benefits as perceived by consumers yield larger estimates but are not suitable for inclusion in a measure of market and near-market production. ### Globalization: geographic boundaries and firm relocations - GDP is defined as production occurring within a country’s economic territory; digitalization has made it easy to fragment global production and separate locations of production and consumption of services; intellectual property assets are easily redomiciled. - Relocations by MNEs for tax reasons can affect reported location of production, and meaning of BOP and IIP statistics; relocations may be accomplished by moving headquarters, implementing holding company structures, or selling intellectual property and other moveable assets to overseas affiliates. - Revisions to the estimates of the 2015 GDP, BOP, and IIP of Ireland highlighted potential effects of relocations: - Yearly GDP growth revised from 7.8 percent to 26.3 percent. - Goods exports revised up by €50 billion. - Net IIP revised from –€150 billion to –€532 billion. - In cases where relocations cause GDP, BOP and IIP data to give an incomplete picture, dissemination of additional detail on MNE transactions and alternative approaches may help. - Ireland developed a modified concept of gross national income that excludes factor income of redomiciled companies and depreciation of relocated assets, and a modified current account balance; modified concepts must be well-documented and provide meaningful insight. - Effects of MNE relocations on measurement of digital output and digital trade vary; output of digital products may be understated in high-tax countries hosting significant operations of digital MNEs. - Apportionment technique example: Guvenen et al. (2017) apportion worldwide income of U.S.-headquartered MNEs to locations where they have operations based on labor inputs and sales to unaffiliated entities, implying: - an increase of 1.7 percent in the estimate of U.S. GDP level in 2012. - U.S. productivity growth increases by 0.1 percentage point in 1994–2004, and 0.25 percentage points in 2004–2008, with no change thereafter. - The upward revision to U.S. GDP could imply an upward revision to net exports of goods and services. - These estimates include all industries; digital enterprises may account for less than half of the total. ### Productivity measurement and the digital economy - Productivity measures changes in the ability of producers to transform inputs into output; only market producers are within scope. - Overstated deflators for ICT products have been identified as causing underestimation of real output and hence productivity. - Nonmarket output is outside standard productivity measures; digitalization has improved households’ ability to produce nonmarket services for own consumption, increasing welfare; measures of these welfare gains are hampered by lack of appropriate time-use data. - Productivity is calculated as growth of market output (as measured in GDP) less growth of inputs (labor, or for TFP, labor and capital combined). - Adjustments to real GDP to better capture output of digital products have a slightly magnified effect on growth rate of aggregate labor productivity because productivity’s base includes only market production. - For TFP, adjustments to output of fixed capital goods, such as computers, may have virtually no effect because capital is also an input. - Context: sharp slowdown in productivity in most AEs that has persisted since the global financial crisis. - Some attribute the slowdown to mismeasurement of the digital economy. - Growth of the market output of the digital sector probably is underestimated because measures of ICT prices underadjust for quality improvements, but ICT output is too small for its measurement to have much impact on aggregate productivity estimates in most AEs. - The slowdown in the productivity growth rate in AEs is more than 1 percentage point, and almost 2 percentage points in some cases. - Underestimation of labor productivity growth caused by underadjustment for quality change in ICT goods and services is generally under 0.3 percentage points—or less, if the mismeasurement before slowdown began is considered. - Effects on TFP are likely to be smaller. - Additional quantitative context: - Output of ICT goods and services is about 6 percent of U.S. GDP in Table 1. - Almost half of this is software, whose growth rate appears to be underestimated only slightly, perhaps by 3 percentage points or less. - Quality improvement in computers and semiconductors seems to be underestimated, perhaps by as much as 10 percentage points, but their weight in the market output covered by the productivity calculations is just 0.5 percent. *Source: 022818measuringdigitaleconomy - 11. The relevant weights for gauging sensitivity of estimates of GDP growth to mismeasurement of the digital sector (PDF chapter/section).* ### 35. Productivity was also underestimated before the slowdown began. Nevertheless, a small ### 022818measuringdigitaleconomy - 35. Productivity was also underestimated before the slowdown began. Nevertheless, a small ### PRICE INDEX COMPILATION CHALLENGES AND STATE OF PLAY - Four challenges posed by the digital economy for price indexes used to deflate digital GDP components and measure inflation: - Capturing the fast-changing quality of digital goods and services. - Handling introductions of new products. - Measuring e-commerce. - Measuring the sharing economy. ### A. Quality Adjustment and Price Indexes - Conceptual point: - Quality improvements represent increases in real output; price indexes must avoid mistaking quality change for price change by adjusting prices of different models or using techniques that avoid cross-model comparisons. - The “matched models” technique excludes non-matching observations; it assumes quality change is reflected in the new-versus-old price differential when models are sold side-by-side. - Illustration (iPhone example, Table 3): - iPhone 6s 32GB transactions price: 2016 = 550; 2017 = 450. - iPhone 7 32GB transactions price: 2017 = 650. - Quality adjustment (iPhone 7 vs iPhone 6s): 200. - Quality-adjusted iPhone 7 price: 450. - Quality-adjusted Price Index for iPhones: 2016 = 100; 2017 = 81.8. - iPhone Nominal Output, Western hemisphere ($ US billions): 2016 = 54.9; 2017 = 60.5. - Nominal Value Index: 2016 = 100; 2017 = 110.2. - iPhone Sales, in constant prices of 2016: 2016 = 54.9; 2017 = 73.9. - Volume index: 2016 = 100; 2017 = 134.7. - Volume index with no quality adjustment (90.6 million in 2017/ 85.1 million in 2016): 2016 = 100; 2017 = 93.2. - Result highlighted: In 2017, output of iPhones is 10.2 percent in nominal terms while its real output growth is 34.7 percent after deflating by the quality-adjusted price index 81.8. - Source for illustration: Staff calculations based on prices observed in April 2017 and Apple’s 2016 Annual Report. - New goods: - Truly novel goods can be incomparable to existing goods even after quality adjustment; welfare gains from such appearances (e.g., smartphones, online platforms) are difficult to quantify and were concentrated in a relatively short period of rapid uptake. - Delays in bringing new goods into the index basket can cause overstatement of price change; current international guidelines allow updating of the basket structure every 5 years. - Research directions: innovative quality adjustment procedures, pricing by function, faster basket updates, use of new data sources. - Survey and compilation practices: - IMF-OECD 2017 survey: 43 responses, 33 from OECD countries. - Just three countries make no adjustments for quality changes in ICT and high-tech products. - For products with regular model turnover (automobiles, computers), most compilers use explicit quality adjustment procedures; quality adjustment for software is particularly challenging. - Compilers in AEs update CPI variety samples annually and make interim sample updates; PPI sample updating practices vary. - Sample rotation practice: entire sample replaced, overlap month pricing; when new models enter during rotations, price change may be overstated. - Research finding: neglected quality improvements in models entering during sample rotations caused price growth to be overestimated by 2 percentage points for televisions (Moulton, Moses and Lafleur, 1998). - Magnitude and cross-country variation: - Combined effects of underestimation of quality change in digital products on consumer inflation in an average OECD economy may be around 0.3 percentage points (see Box 2). - Price indexes for ICT equipment, software, and telecom services vary greatly across countries (1994=1.00 or 2002=1.00 bases in figures cited). - Ahmad et al. (2017) adjusted growth rates over 2010–15: adjusted indexes for Germany and France fall more than 4 percent per year; Belgium, Italy, and the United Kingdom rise or are flat; remaining countries fall by 2 to 3 percent per year. - Telecom services example: Abdirahman et al. (2017) find a 7-percentage point overstatement of the price growth rate in 2010–2015 in the United Kingdom. - Internet access and online content: - If Internet access is treated as a product, growth in online content and faster connection speeds represent quality improvements; however, increases in hacking, trafficking of stolen information, and ransomware attacks imply quality declines and increased cybersecurity expenditures. - Recommendations for NSOs: - Quality-adjust prices of key digital products drawing on other countries’ experience. - Consider innovations in data sources, data collection, and index calculation to incorporate new varieties and suppliers quickly. - Maintain up-to-date samples of varieties and suppliers to align deflators with national accounts aggregates. - Foster close collaboration between price statisticians and national accounts compilers. - Follow recommendations of the Voorburg Group and the Ottawa Group as a basis. ### Box 2 — Quantified effects on inflation and cost of living (summary of recalibrations) - Three kinds of effects on cost of living potentially not captured in CPI and household consumption deflator: - Unmeasured quality improvements in ICT goods and services. - New free and inexpensive digital products replacing non-digital products. - E-commerce expanding variety and enabling better variety selection through online search/results. - Reinsdorf and Schreyer (2017) calibration assumptions (using OECD PPP program weights): - Assume ICT equipment and telecommunication services indexes overstate price growth by 5 percentage points due to underestimation of quality improvements. - Assume other goods incorporating digital technology (televisions, automobiles) overstate price growth by 2 percentage points. - Based on 2015 consumption basket weights, the deflator for household consumption would overestimate inflation by 0.28 percentage points from these effects. - For categories with free or inexpensive digital replacements (taxis, music and video recordings, newspapers, film developing), assume overstatement of cost-of-living change by 5 percentage points; for weakly affected categories (books, postal services, travel formerly via travel agents), assume 2 percentage points. - Effect on the household consumption deflator is 0.18 percentage points with 2005 weights, falling to 0.11 percentage points with 2015 weights. - Marginal benefits of expanded variety and better variety selection could be 0.3 percentage points overstated at product level, implying a 0.06 percentage point impact on the household consumption deflator. - Combined 2015 bounds for the three effects give an upper bound of 0.45 percentage points for the combined overstatement of the rate of change in the cost of living. - Context: The productivity slowdown has reduced growth by more than 1 percentage point in the AEs. ### B. Coverage of E-Commerce and the Sharing Economy - E-commerce adoption statistics and scope: - U.K. ONS data: non-gasoline retailers in the United Kingdom in 2017 made more than 16 percent of their sales online. - Eurostat (EU, 2016): 66 percent of household Internet users made online purchases; businesses with 10 or more employees reported sales over the Internet to consumers equal to 2.7 percent of turnover and all other e-commerce sales equal to 15.6 percent of turnover. - Price implications and measurement gaps: - Online prices may be lower; Cavallo (2017) finds Amazon prices are, on average, 5 percent lower than prices at offline stores. - If substitution to online shopping reduced average price paid, CPI may not fully capture the cost-of-living effect due to under-representation of online prices in CPI samples and treatment of outlet price differentials as quality changes by about half of IMF-OECD survey respondents. - Adobe Digital Price Index (DPI) vs U.S. CPI over 2014–2017: DPIs fall by 1 percent per year, on average, relative to corresponding CPIs (Goolsbee and Klenow, 2018); procedural differences and atypical years may contribute. - Billion Prices Project data generally show similar rates of change in online and offline prices (Cavallo and Rigobon, 2016). - Survey coverage of e-commerce in price indexes (IMF-OECD survey, sample of 43 countries): - Domestic e-commerce included in CPI: 30 (69.8). - Domestic e-commerce included in PPI: 13 (35.1). - Cross-border e-commerce included in CPI: 12 (27.9). - Cross-border e-commerce included in PPI: 7 (18.9). - Regional breakdown (counts and percentages in parenthesis): - Africa: CPI = 1 (16.7); PPI = 0 (0.0); cross-border CPI = 0 (0.0); cross-border PPI = 0 (0.0). - Asia-Pacific: CPI = 4 (66.7); PPI = 4 (80.0); cross-border CPI = 2 (33.3); cross-border PPI = 3 (60.0). - Europe: CPI = 21 (91.3); PPI = 7 (33.3); cross-border CPI = 8 (34.8); cross-border PPI = 3 (14.3). - Middle East & Central Asia: CPI = 1 (50.0); PPI = 0 (0.0); cross-border CPI = 0 (0.0); cross-border PPI = 0 (0.0). - Western Hemisphere: CPI = 3 (50.0); PPI = 2 (33.3); cross-border CPI = 2 (33.3); cross-border PPI = 1 (16.7). - Overall finding: About 70 percent of respondents include e-commerce prices in their CPI; PPI coverage is limited. - Gaps exist for products frequently purchased online (clothing and footwear); lags in reflecting fast-changing purchasing patterns contribute to under-representation. - Implication and priority: - Adequate coverage of e-commerce in price samples is important because online prices may have different growth rates; enhanced representation and timely sample updating are needed. *International Monetary Fund — content from the specified chapter/section of the source PDF.* ### 51. Sharing economy suppliers should be included in the CPI with weights that reflect their ### 51. Sharing economy suppliers should be included in the CPI with weights that reflect their importance in consumers’ spending patterns. ### Inclusion of sharing-economy suppliers in CPI and welfare measurement - Sharing economy suppliers should be included in the CPI with weights that reflect their importance in consumers’ spending patterns. - Where substitution to a lower priced replacement product in the sharing economy has been significant, estimates of the effect on the cost of living are important for understanding welfare change and may improve CPI accuracy. - If the replacement product from the sharing economy is a close substitute for the original product (example given: Uber and taxis), it may be reasonable to let CPI reflect the change in the average price paid for the service. ### Progress on including sharing-economy prices in price indices - Only three respondents to the IMF-OECD survey (Australia, Germany and the United States) include these prices in their CPI, and none include them in the PPI. - Measurement error in indicators of price and output growth depends partly on the relative size of the sharing economy, which still tends to be small. - Many survey respondents regarded the sharing economy as relatively unimportant in their country. ### National accounts compilation challenges — A. E-Commerce and Free Products from Online Platforms - 2016 OECD survey (29 responses) showed slow progress in developing estimates of the digital sector. - 2017 IMF extension surveyed some non-OECD countries (11 responses); many indicated measuring the digital sector was not a priority. Common impediments: lack of resources and unavailability of source data. - Malaysia is developing an ICT satellite account that includes online platforms. - Only a third of countries responding to OECD and IMF surveys collect data on online purchases; just five collect separate data on cross-border e-commerce transactions. - Ghana, India, Jamaica and Malaysia reported including data on e-commerce in national accounts compilation. - Proposals to impute viewers’ consumption of free online media funded by advertising and collection of data are under discussion; just one country has developed experimental estimates. - Only eight respondents agreed that free products funded by advertising should be included in household consumption or in a new category of final consumption. - No country has data to impute production by volunteers of free online content/media, or of free software, nor information on who consumes these free products. - Lack of coverage of open source (free) software is a statistical concern because inputs of open source software can affect measurement of TFP and of commercial software production. - GDP compilers often base part of software production estimates on input costs, including earnings of software coders; commercial enterprises may pay coders to write open source software as a way of generating sales of support services or premium upgrades (the “freemium” business model). ### National accounts compilation challenges — B. The Sharing Economy - Sharing economy suppliers of short-term property rental services or labor services may be informal (unregistered and untaxed). National accounts procedures for informal rental and labor activity tend to reflect pre-sharing-economy conditions. - Household surveys help measure labor activity, but respondents often neglect to report sharing-economy work unless questionnaires include explicit questions (Abraham et al., 2017). - Countries are progressing on bringing platform-enabled rentals and labor services into tax, regulatory and reporting regimes, which should help GDP measurement. - Platforms such as Airbnb have enabled rapid growth of short-term rentals in some economies, particularly those with tourism, suitable housing stock, and favorable legal environments. - Just six of the OECD-survey countries capture rentals of property through digital intermediaries; in the IMF survey, only India reported capturing these rentals. - Information for estimates may come from tax data or be collected directly from the digital intermediary. - Relative to GDP, scale of rental activity is limited; press reports of Airbnb’s revenue forecast for 2017 of $2.8 billion suggest worldwide host revenue of perhaps $30 billion. - GDP already includes an imputation for services of owner-occupied dwellings that captures part of the value of peer-to-peer rentals. - For labor services (e.g., Uber): in about half the countries where they are legal, GDP compilers include sharing-economy transportation services in estimates of self-employment income based on tax data (8 countries) or labor force surveys (7 countries, with some using both). - India, Malaysia and Ghana capture some of these earnings through labor force surveys. - Intermediation fees retained by platforms are challenging to measure; only five countries in the survey reported that they currently capture these fees. - Often intermediation services are supplied by a foreign platform that does not report cross-border transactions, or available data lack detail on the intermediation fee component of revenue. ### National accounts compilation challenges — C. Lags and Data Gap Concerns - International guidelines recommend updating the benchmark year of the national accounts at least every five years; rapid online platform growth may make five-year intervals underestimate sector size. - Composition of available deflators often lags composition of aggregates they deflate; varieties and detailed weights in PPIs or CPIs likely reflect previous-year patterns while national accounts aggregates reflect current patterns. - Example: a deflator might contain only laptops while tablets become predominant in spending; sharing-economy expenditures may enter GDP compilation before they enter CPI or PPI. - Better correspondence between deflators and national accounts aggregates composition would improve growth measurement accuracy in the rapidly changing digital sector. - Data users need more extensive and more granular statistics on scale and structure of digital activity; alternative classifications may be required (example: employment in e-commerce retailing may need to include establishments classified in warehousing). - Estimates of scale and relative importance of digital activities are needed to analyze potential mismeasurement of growth and productivity. - Ahmad and Ribarsky (2017) discuss work on guidelines for a digital economy satellite account. ### Compilation challenges and state of play in external sector statistics — A. Digital Trade - Digital trade is growing in importance, raising measurement concerns and data dissemination needs. - International organizations involved in trade statistics have initiatives, including a project on cross-border e-commerce and collaboration on guidelines for measuring and classifying digital trade. - A preliminary measurement framework for digital trade considers multiple dimensions and types of transactions. - Data sources and methods for capturing new business models (Uber, Airbnb, Facebook, Spotify) are being developed. - Digital trade includes cross-border transactions that are digitally ordered, platform-enabled, or digitally delivered. - Digitally-delivered products include software, media and cross-border data flows (e.g., advertisements). - UNCTAD’s guidelines define “ICT-enabled services” as comprising digitally-delivered services consumed remotely. - In 2014, ICT services and (potentially) ICT-enabled services accounted for: - 54 percent of overall services exports and 48 percent of services imports in the United States. - 56 percent of services exports to non-EU countries, and 52 percent of services imports from non-EU countries in the EU. - In emerging market and developing economies in 2014, ICT services and (potentially) ICT-enabled services accounted for 33 percent of services exports and 27 percent of services imports. - China and India play leading roles. - Digitally-delivered services raise conceptual and practical questions, including boundaries of what is included and treatment of free media funded by advertising, which may involve international flows of unpriced data. - Platforms often produce services with unpriced cross-border data flows as intermediate inputs; little progress has been made on measuring these data flows. - 2017 OECD and IMF survey of BOP compilers (74 countries) highlights practical measurement challenges and gaps in source data. - Digital ordering, digital payments, and digital delivery have fostered growth of small transactions below customs reporting thresholds; most countries have reporting thresholds but sizes and adjustment practices vary. - More than half of OECD countries include an estimate of below-the-threshold transactions in IMTS; a minority of non-OECD countries do so. - The largest estimate of the share of below-the-threshold trade among IMF survey respondents was about 15 percent. - U.S. estimates were 2.2 percent of goods imports and 0.8 percent of goods exports in 2015. - Example growth: in Russia, share of e-commerce in imports grew from 2.5 percent in 2016 to 4.2 percent in the first quarter of 2017. - Few countries have separate estimates of cross-border e-commerce. - Digital downloads are covered by source data in most OECD countries and almost half of non-OECD countries, though may not be separately identifiable. - Eleven countries indicated household consumption of cross-border services could potentially be captured through credit card data; other sources including administrative data are under consideration. - Many respondents viewed trade in digitally-delivered services as under-reported, particularly imports; Luxembourg’s services exports to EU countries substantially exceed the imports recorded by trading partners (example: Spotify capture issues referenced). - Only three surveyed countries are researching methods to measure trade in sharing-economy services. - Almost a third of OECD countries, and almost a quarter of non-OECD countries, reported sharing-economy services are implicitly included in cross-border trade totals. - Intermediation services of non-resident platforms are rarely measured: only 15 percent of OECD countries and 6 percent of non-OECD countries can identify payments to nonresident digital intermediaries. - Foreign-owned digital intermediaries are generally hard to identify on business registers; in three-quarters of OECD countries, and nearly 90 percent of non-OECD countries, they cannot be easily identified. - Cross-border data flows that do not result in monetary transactions may indirectly support revenue-generating activities (e.g., social networking platforms with advertising revenue); none of the OECD survey respondents have researched measures of unpriced data flows; most opposed including imputed values of cross-border data flows in balance of payments statistics for conceptual and practical reasons. ### Compilation challenges and state of play in external sector statistics — B. Digital Payments and Measurement of Cross-Border Remittances - Standard methods for capturing cross-border payments in BOP statistics may omit some payments using digital technologies; effects could be important for international remittances. - Table 5. Global Totals of Remittance Receipts and Payments (Billions of U.S. Dollars) - Credit: 2009 397.3; 2010 424.2; 2011 472.6; 2012 496.1; 2013 525.4; 2014 555.9; 2015 550.8 - Debit: 2009 303.3; 2010 302.9; 2011 340.6; 2012 356.3; 2013 390.4; 2014 400.5; 2015 386.0 - Discrepancy: 2009 94.0; 2010 121.3; 2011 132.0; 2012 139.8; 2013 135.0; 2014 155.4; 2015 164.8 - Households increasingly use digital platforms to make remittances; mobile money has significant share of cross-border transfers, especially between neighboring countries where available. - Mobile money may offer savings on transactions costs for remittances exceeding 50 percent (GSMA, 2016a), and are available even in remote areas. - Many corridors for mobile money remittances were already active as of 2015. - New digital channels (mobile money, online platforms, Bitcoin) may exacerbate gaps in source data for estimating remittances. - Common data sources for remittances: reports on cross-border transactions from banks and money transfer operators, and household surveys. - World Bank central bank survey (2008–2009) found just 43 percent of remittance-receiving countries collect information on remittances through informal channels (cash brought in pockets and unregistered agents). - Some international mobile money transfers are probably captured when they pass through banks partnered with mobile money operators; others are made directly by mobile phone operators (Bank of Uganda, 2017). - In countries where mobile money is widely used for remittances (example countries: Uganda, Kenya, Philippines), no data are collected from telecommunication companies on cross-border remittances executed via their networks, and no adjustments are made for remittances through such channels. - Data accuracy could improve as digital transfers replace informal mechanisms and digital transfer platforms are brought into regulatory and reporting regimes. - Measuring mobile money remittances is an area of research; proposals exist to address data gaps by surveying telecommunication companies involved in digital transfers. Surveys of telecommunication companies should not be expensive to conduct. *Source: INTERNATIONAL MONETARY FUND (excerpt).* ### 82. The financial sector has long been an intensive user of digital technology, with the ### 82. The financial sector has long been an intensive user of digital technology, with the recent digitalization of financial services being termed “fintech.” ### A. Marketplace Lending Platforms - Marketplace lending platforms facilitate peer-to-peer (P2P) credit transactions or lend their own funds. - P2P platforms primarily match borrowers with individual investors, but may also allow banks and other institutional investors to invest in loans. - Some marketplace lenders have arrangements with a bank to securitize their loans (example: SoFi and Prosper). - Geographic concentration and access: - Marketplace lenders operate in many countries, with large lending volumes in China, the United States, and the United Kingdom. - They have improved access to credit and lowered the cost of credit for SMEs and households by operating in underserved areas and using credit scoring models that consider non-traditional information. - Measurement and classification issues: - Because they are less regulated than conventional financial intermediaries, marketplace lenders may not be subject to reporting requirements, delaying inclusion in MFS data and requiring direct surveys by statisticians. - MFS compilation guidelines classify P2P lending platforms as financial auxiliaries because in most cases they do not assume credit risk and rely on fees collected from users for income (IMF, 2016). - Loans made by households with the assistance of these auxiliaries are currently not included in MFS; households’ P2P lending may need to be reported as supplementary information. - Loans by households to SMEs via P2P should be included in the integrated sectoral financial accounts of the SNA. - Marketplace lenders that extend credit from their own funds should be included in the financial sector as financial intermediaries, but the information to do so may be unavailable. ### B. E-Money - Definition: - E-money is defined as a monetary value, represented by a claim on the issuer, that is electronically stored on a card, device or server, and used for payments to third parties. It includes mobile money, widely-accepted pre-paid cards, and web-based products. - Exclusions: digital currencies are not e-money; credit and debit cards are not e-money (no monetary value stored on them); mobile phone payment applications linked to an account at a financial institution are excluded; pre-paid cards usable only at certain stores are excluded. - Role and prevalence: - Although e-money has a low share in the money supply and in payment transactions (Table 6), as an accessible substitute for transferable deposits it plays an important role in financial inclusion. - Mobile money, a form of e-money stored in mobile phone accounts, is widely used in many emerging and developing economies. - As of 2015, the number of mobile money accounts approached or exceeded the number of bank accounts in 21 African countries (Table 7), and mobile money was present in more than 90 countries (Figure 9). - Among the benefits of mobile money for financial inclusion is greater access to financial services for women. - Measurement and data implications: - Accuracy of existing MFS appears largely unaffected by e-money because regulations generally require issuers to mirror outstanding e-money in an escrow account at a regulated financial institution. - Mobile money operators offering non-payment services pose no measurement problems if they partner with a regulated financial service provider covered by reporting requirements or obtain a license that brings them into the regulatory regime. - Mobile money has implications for data dissemination: data are needed to analyze effects on financial inclusion and may create opportunities to improve accuracy of other kinds of statistics on households using provider data. - Selected data points (Table 6: Share of E-money Transactions in Noncash Payments in 2015) - Advanced Economies: - Austria 0.01 - Belgium 0.01 - Cyprus 0.02 - France 0.00 - Germany 0.00 - Greece 0.07 - Ireland 0.02 - Italy 0.21 - Korea, Republic of 0.00 - Malta 0.03 - Netherlands 0.00 - Norway 0.00 - Portugal 0.08 - Singapore 0.24 - Slovenia 0.00 - Sweden 0.00 - Switzerland 0.03 - Emerging and Developing Economies: - Albania 0.02 - Brazil 0.01 - Bulgaria 0.02 - Croatia 0.00 - Dominican Republic 0.01 - Egypt 0.14 - India 0.05 - Maldives 0.04 - Namibia 0.18 - Nigeria 0.94 - Russian Federation 0.13 - Samoa 0.16 - Thailand 0.02 - Selected data points (Table 7: Mobile Money Share of Accounts — Percent of Total Number of Accounts at Mobile Phone Companies and Commercial Banks, data cover 2015 or most recent year available) - Afghanistan 6.04 - Bangladesh 30.60 - Benin 50.95 - Botswana 51.19 - Burkina Faso 47.76 - Cambodia 14.33 - Cameroon 79.62 - Cote d'Ivoire 78.10 - Egypt 12.33 - Ghana 53.31 - Kenya 47.34 - Madagascar 65.21 - Mali 63.61 - Mozambique 42.42 - Namibia 33.13 - Nigeria 13.75 - Pakistan 25.40 - Philippines 19.61 - Rwanda 84.84 - Senegal 65.00 - Tanzania 87.40 - Uganda 81.89 - Vietnam 6.24 - Zambia 65.29 - Zimbabwe 91.51 - Notes on tables: - Total non-cash payments equal the sum of debit card payments, credit card payments, direct debits, credit transfers, checks, and e-money payments. Data cover 2015 or most recent year available. ### C. Digital Currencies - Definition and scope: - A digital currency is a means of payment that only exists electronically. Cryptocurrencies such as Bitcoin are the best-known type of digital currency. - Over 1,300 digital currencies exist. - In early 2017, there were between 2.9 and 5.8 million unique active users of digital currency wallets (Hileman and Rauchs, 2017). - Potential systemic and measurement implications: - Digital currencies have the potential to affect the measurement of liquidity in the financial system if they become widely accepted as a medium of exchange. - Several central banks are investigating issuance of digital currencies, on which data are not reported at present. - Private digital currencies raise measurement issues for financial, macroeconomic and balance of payments statistics because the residency of the holders is unavailable. - Current classification: - Existing digital currencies do not qualify as money in the current MFS framework and are classified as nonfinancial assets in MFS compilation guidelines. - According to the internationally-accepted MFS framework, Bitcoin is not classifiable as money because it is not issued or authorized by a central bank or government, is not widely accepted as a medium of exchange, and exhibits excessive price volatility to be considered a store of value. - If and when digital currencies become widely accepted as a medium of exchange, consideration will be given to their inclusion in a broader measure of liquidity. - Selected market capitalization comparisons: - As of March 2017, Bitcoin’s total market capitalization was US$25 billion, compared with US$1.5 trillion of U.S. dollar currency in circulation. - In February 2018 Bitcoin’s capitalization was almost $150 billion. ### Policy Implications and Recommendations - High-level findings: - Slow productivity growth since the financial crisis is a genuine phenomenon requiring policy responses, not merely a measurement error. - Where research and data exist (for example, in the United States), underestimation of digital sector output could subtract 0.3 percentage points from the productivity growth rate, compared to a 1- to 2-percentage point slowdown in productivity growth rates. - Largely symmetric effect on price statistics can yield a slight overestimation of inflation due to under-adjustment for quality change and lags in including new digital products and suppliers. - Digitalization has exacerbated weaknesses in compilation methods and created new data needs across macroeconomic statistics; there may be a need to consider adjustments to the GDP production boundary and alternative treatments in national accounts and BOP statistics. - Institutional roles: - The Fund is positioned to promote international cooperation and research on definitions, classifications and measurement techniques under its articles of agreement (Art. VIII, Section 5). - International organizations should update classification systems for digital activities and products and develop guidelines on measuring digital transactions. - The Fund should be ready to respond to requests from member countries for technical assistance on measurement of the digital economy. - Recommendations (general): a) Improve coverage of digital platforms and services linked to digital platforms in the main classification system; and b) Develop aggregate classifications covering the digital sector, digital products and digital transactions. - Recommendations on resources and data access: - National statistical offices require additional resources and better access to source data to implement compilation improvements and disseminate additional indicators, including possible new indicators of welfare and nonmarket production linked to digitalization. - Declining survey response rates and emerging new sectors make it essential to expand use of administrative data and “Big Data.” - Governments should be encouraged to share data needed for statistics; national and international organizations should facilitate access to Big Data through partnerships with global firms. - Recommendations: a) Endow national statistical offices with sufficient resources to measure digital products and to develop indicators of welfare effects of digitalization; and b) Ease access by national statistical compilers to administrative data, and Big Data, promoting sharing administrative data with statistical agencies, as well as partnerships between the private and public sectors, including international organizations. - Recommendations for price statistics: - Main compilation challenges: improving quality adjustment procedures for ICT goods and services; timely inclusion of new digital product varieties and suppliers in detailed indexes; timely inclusion of new digital products in the basket and weighting structures of the high-level index. - Recommendations: a) Statistical agencies should focus on quality-adjusting a selective list of products drawing on quality adjustment work of other countries; b) International organizations should develop compilation approaches fit-for-use by compilers facing severe resource and data sources constraints; and c) Statistical agencies should consider innovations in data sourcing and collection, and processing to include new digital products in index compilation as soon as they become important. - Recommendations for national accounts: - Key compilation challenges: deflation and complete coverage of digital platforms and platform-enabled activity. - Recommendation: a) National accounts compilers and price statisticians should work collaboratively to align the composition of the deflators for digital products with the composition of the aggregates that need to be deflated by ensuring that the deflators reflect the current mix and sourcing of digital products and, where possible, by using datasets containing prices and quantities to simultaneously calculate deflators and nominal values. - Recommendations for external sector statistics: - Measurement challenges include growth of small transactions and of cross-border services and payments using digital platforms. - Recommendations: a) Statistical agencies should update assumptions concerning small transactions facilitated by digital ordering and digital delivery of services; b) Enhance collection of information on cross-border services provided by, or through, online platforms; and c) Develop methods for estimating international payments via new kinds of digital channels, such as remittances via mobile money. - Recommendations for monetary and financial statistics: - In the future the new liquidity and credit generated by fintech could become important. - Recommendations: a) Add marketplace lending platforms that lend their own funds to credit statistics, and report supplementary data on peer-to-peer lending; and b) Investigate methods for compiling statistics on digital currencies. *International Monetary Fund — Measuring the Digital Economy (selected section).* ### References ### References ### Bibliographic list - Comprehensive bibliography of works cited on measuring the digital economy, including academic articles, working papers, IMF Staff Discussion Notes, national statistical office reports, BIS/CPMI publications, OECD papers, GSMA reports, NBER working papers, and presentations at the 5th IMF Statistical Forum and other conferences. - Representative entries (preserve original citations and terminology exactly as presented): - Abdirahman, Mo, Diane Coyle, Richard Heys, and Will Stewart, 2017, A Comparison of Approaches to Deflating Telecoms Services Output. Presented at the 5th IMF Statistical Forum. http://www.imf.org/~/media/Files/Conferences/2017-stats-forum/session-6-heys.ashx?la=en - Ahmad, Nadim, and Jennifer Ribarsky, 2017, Issue Paper on a Proposed Framework for a Satellite Account for Measuring the Digital Economy. http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=STD/CSSP/WPNA(2017)10&docLanguage=En. - Byrne, David and Carol Corrado, 2015, “Prices for communications equipment: Rewriting the record”. FEDS Working Paper 2015-069 (September), Federal Reserve Board, Washington, D.C. http://dx.doi.org/10.17016/FEDS.2015.069. - Committee on Payments and Market Infrastructure (CPMI), 2016, “Statistics on payment, clearing and settlement systems in the CPMI countries”, Bank of International Settlements, Basel, 2016, https://www.bis.org/cpmi/publ/d152.htm. - European Commission, International Monetary Fund, Organisation for Economic Co-operation and Development, United Nations and World Bank, 2009, System of National Accounts 2008 (New York: United Nations). https://unstats.un.org/unsd/nationalaccount/docs/sna2008.pdf. - GSMA, 2016a, Driving a price revolution: Mobile money in international remittances. GSM Association, London. https://www.gsma.com/mobilefordevelopment/wp-content/uploads/2016/10/2016_GSMA_Driving-a-price-revolution-Mobile-money-in-international-remittances.pdf. - Hulten, Charles and Leonard Nakamura, 2017, Accounting for Growth in the Age of the Internet: The Importance of Output-Saving Technical Change. NBER Working Paper No. 23315. http://www.nber.org/papers/w23315 - Reinsdorf, Marshall and Paul Schreyer, 2017, Measuring Consumer Inflation in a Digital Economy. Presented at the 5th IMF Statistical Forum. http://www.imf.org/~/media/Files/Conferences/2017-stats-forum/session-1-schreyer-and-reinsdorf.ashx?la=en. - Torslov, Thomas, Ludvig Wier, and Gabriel Zucman, 2017, €600 Billion and Counting: Why High-Tax Countries Let Tax Havens Flourish. http://gabriel-zucman.eu/files/TWZ2017, pdf - The references include multiple repeated and related entries on themes such as: deflation and price measurement for telecoms and communications equipment; measurement of free digital services and consumer surplus; measuring sharing and collaborative economies; measuring digital trade and ICT-enabled services; mobile money and digital finance; big data and statistical implications; and the System of National Accounts 2008 and classification standards (ISIC Rev. 4, CPC v2.1). ### Thematic emphasis in cited literature - Measurement methods and challenges: - Deflators and price measurement for telecoms, communications equipment, semiconductors, and cloud computing. - Approaches to valuing “free” digital services and consumer surplus (e.g., Brynjolfsson et al., Nakamura et al., Cavallo). - Satellite accounts and household production measurement (e.g., Ahmad & Ribarsky; Office for National Statistics household satellite accounts). - Digital finance and payments: - Mobile money definitions, deployment, and remittance pricing (GSMA; Bank of Uganda; CPMI). - Fintech lending, peer-to-peer lending, and implications for financial inclusion (Jagtiani & Lemieux; Positive Planet; World Economic Forum; Accenture). - Cryptocurrency benchmarking and electronic money regulation (Hileman & Rauchs; Bossone). - Trade, multinational enterprises, and national accounts: - ICT-enabled services trade measurement and digital trade frameworks (Fortanier & Lopez Gonzalez; UNCTAD; OECD). - Offshore profit shifting and implications for productivity and current account measurement (Guvenen et al.; Rassier; Torslov, Wier, & Zucman). - Big data and new data sources for mapping digital businesses and internet economy measurement (Hammer et al.; Nathan & Rosso; Oostrom et al.; Muenchen). ### Annex I. Link Between Consumption and Welfare Growth - Key points: - The level of consumption, measured by price times quantity, understates welfare because it excludes the consumer surplus. In the diagram described: - Initial demand implies a quantity consumed at price p of q0, making consumption equal to the area of rectangle c. - Consumer surplus is the area under the demand curve above the price line, the triangle labeled s. - Welfare is measured by the area under the demand curve out to q0, c+s. - The growth of real consumption equals (or approximates) the welfare growth. Assuming income growth shifts the demand curve right so quantity becomes q1 and price is constant: - Nominal consumption growth, given by (c+Δc)/c, equals real consumption growth, q1/q0. - Welfare growth, given by (c+Δc+s+Δs)/(c+s), also equals real consumption growth. - The weights used to calculate aggregate growth of real consumption are based on prices as the measure of value. These weights allow aggregate growth to approximate the welfare growth. *International Monetary Fund — References section.* --- _Source: https://www.imf.org/-/media/files/publications/pp/2018/022818measuringdigitaleconomy.pdf_