U.S. Consumers’ Adoption and Use of Bitcoin and other Virtual Currencies Scot t Sch u h a n d Oz Sh y P r e s e n t e d a t D N B P a ym e n t s Co n fe r e n ce “R e t a il p a y m e n t s : m a p p in g o u t t h e r o a d a h e a d ” Am s t e r d a m , Ap r i l 2 1, 2 0 16 * The view s are those of the author and do not reflect the view s of the Federal Reserve Bank of Boston or the Federal Reserve System .
Overview Context and motivation Definitions and concepts Survey of Consumer Payment Choice (SCPC) Survey Results Econometric Results Awareness and familiarity Adoption and use for payments Value of holdings (VC and cash) 2
Context and motivation 3
Bitcoin price and search interest are correlated Correlation: 0.80 Source : blockchain.info and Google Trends. Bitcoin search interest is based on Google Trends in the US, scaled to overlay with prices. 4
Bitcoin learning was followed by rapid growth in many other virtual currency competitors Source : mapofcoins.com and Google Trends. Number of virtual currencies include 112 defunct coins. Cumulative Bitcoin search interest is based on Google Trends in the US, scaled to overlay with number of coins. 5
Definitions and concepts 6
Clear terminology about complex concepts is crucial for surveys and consumer cognition Bitcoin = “A peer-to-peer electronic cash system” Digital = electronic, esp. computerized technology (Internet or cellular); perhaps binary (0/ 1) representation Virtual = digital + never physical US dollar is digital (debit, ACH) and physical (currency) Network (peer-to-peer) – closed loop versus open loop Crypt0 = security based on modern encryption algorithms Used to secure transactions and to control the creation of new virtual currency units. Money or currency Ownership – public (“sovereign”) versus private Value – asset-backed (gold) versus “fiat” (promise) e-Money = public money stored electronically on a physical device (card) 7
ECB classified “ virtual currency ” as a form of money but did not highlight private ownership More VC details • Netw ork access : closed or open loop • Security : Internet or cryptography (“math-based”) • Bitcoin is open loop, cryptographic (by far the largest of about 700 of these) BIS (20 15): used “d igita l currency ” but v iew ed “v irtua l currency ” a s equiv a lent. Source : “Virtual Currency Schemes”, ECB 2012. 8
Summary of key literature Origin of Bitcoin Nakamoto (2008) Bitcoin/ VC as technology and m oney ECB (2012), Velde (2013), Greene and Shy (2014), Lo and Wang (2014), Böhme et al (2015), BIS (2015), Badev and Chen (2014) Bitcoin/ VC as an asset and pricing Glaser et al (2014), Hencic and Gouriéroux (2014), Donier and Bouchard (2015), Dywer (2015), Bolt and van Oordt (2015) Network econom ics of the Bitcoin/ VC industry Gandal and Halaburda (2014), Möser and Böhme (2015) Acceptance and use of Bitcoin/ VC MERCHANTS: Polasik et al (2015) CONSUMERS: Christin (2013) , Tsanidis et al (2015), Saito (2015), Coinbase (2015), MIT Digital Currency Project (ongoing) 9
Survey of Consumer Payment Choice (SCPC) 10
2014–2015 SCPC characteristics 11
Survey Results 12
Less than half of U.S. consumers are aware of VC and most of them say their familiarity is low 13
Adoption of virtual currency is low but appears to be growing relatively fast like new technologies 14
Reasons for adopting virtual currency are primarily related to payments activity 31.4 Buy goods and services 3.7 14.2 Interested in new tech 10.9 11.8 Make remittances 32.4 11.1 Investment 9.4 Primary 10.7 Other 5.6 Secondary 8.2 Pay anonymously 15.1 Primary reason payments related 7.5 Uses blockchain to prevent… 7.9 >50% 5.1 Do not trust gov't or USD 2.0 0.0 Do not trust banks 13.0 Percentage of virtual currency adopters 2015 Survey of Consumer Payment Choice Preliminary result, unofficial and subject to revision 15
Reasons for not adopting virtual currency are diverse and related to all aspects of VC It is not easy to acquire or use. 18.7 I do not understand the technology. 17.7 Not accepted for payment very often. 16.0 The U.S. dollar value of the virtual 15.2 currency varies too much. My current payment methods meet all of 14.5 my needs. It is not guaranteed by the U.S. 12.8 government. Other 5.1 Percentage of virtual currency nonadopters who are aware of virtual currency 2015 Survey of Consumer Payment Choice Preliminary result, unofficial and subject to revision 16
Diffusion model of Bitcoin awareness shows slow learning and modest outlook for adoption Method : Bass (1969). "A new product grow th for m odel consum er durables". Managem ent Science 15 (5). 17
Bitcoin adopters are much more optimistic about future prices ($/ coin) than non-adopters… . Exchange rate expectations VC Non-adopters 1 year from now 58 61% Bitcoin adopters VC adopters (no Bitcoin) 34% other VC adopters Bitcoin adopters 17% nonadopters 45 33 28 28 20 15 14 14 12 11 8 6 5 3 Decrease a Decrease Stay about Increase Increase a lot some the same some lot 18
… .but Bitcoin adopters have less biased price forecasts than pessimistic non-adopters 19
Nearly half of adopters use their VC to make payments for goods and services 20
The actual reported data on VC holdings show signs of significant measurement errors 21
Heavily screened/ imputed data show median VC holdings of <$1,000 per adopter 22
Econometric Results 23
Underpinnings of econometric models Awareness and fam iliarity Models of information diffusion (Bass 1969); financial literacy Dem and for Bitcoin/ VC OPTION 1 – Adoption (extensive margin) To own or not to own? (Literature on consumer demand for payments) OPTION 2 – Amount of holdings (intensive margin) Extension of money demand? M1 – Baumol (1952), Tobin (1956) etc.; cash– Briglevics and Schuh (2013) New type of asset demand, pricing? Expectations of future price ($ exchange rate) – Bolt & van Oordt (2015) Use of Bitcoin/ VC Two-step Heckman selection models (e.g., Schuh and Stavins 2010) Prelim inary results are single-equation m odels Adoption/ use sim ultaneous equations Application of KRSS (2016) left for future research… 24
Qualitative summary of regression results for Bitcoin only: Demographic variables Awareness Adoption Use Holdings Age - - Education (college/ postgrad) + - Household size - + White + - Male + + Income + - Number of other PIs adopted + + + PayPal adoption + + + Google Wallet adoption + + + Cash share Debit share - - GPR card adoption - Networth/ income + + - 25
Qualitative summary of regression results for Bitcoin only: Additional variables Awareness Adoption Use Holdings Online retail share na na na Use of internet bank + na na na Expectations (week) na + na Expectations (month - week) na + na Expectations (year - month) na + na Interested in new technology na na omitted na Buy Goods and Services na na + na Remittances na na + na Anonymity na na + na Investment na na + na Blockchain technology na na + na No trust in banks na na + na No trust in government na na na Checking account pays interest na na na Checking account interest na na na Savings account pays interest na na na Savings account interest na na na 26
Awareness regressions Regression Description • Marginal effects of logits on: 1. Awareness of Bitcoin 2. Awareness of other VCs • OLS on: 3. Familiarity with Bitcoin, conditional on awareness (1-5 scale) 27
Expectations regressions Regression Description • OLS on: 1. Expectations of Bitcoin prices (week, 1-5 scale) 2. Expectations of Bitcoin prices (month, 1-5 scale) 3. Expectations of Bitcoin prices (year, 1-5 scale) 28
Adoption and Use regressions Regression Description • Marginal effects of logits on: 1. Adoption of Bitcoin 2. Adoption of other VCs 3. Incidence of use of Bitcoin in the past 12 months 4. Incidence of use of other VCs in the past 12 months 29
Holdings regressions Regression Description • OLS on: 1. Log of Bitcoin holdings (best, unscreened) 2. Log of cash holdings 30
Preliminary conclusions Awareness and especially familiarity are relatively low 1. (even among adopters) and growing slowly Adoption is low but growing fast; adopters are very 2. optimistic about future price appreciation Many adopters frequently make payments to 3. merchants (P2B) and people (P2P) Survey results are interesting but imperfect so (much) 4. more survey work and research are needed! 31
Recommend
More recommend
