Cryptocurrencies: Stylized facts on a new investible instrument

• DOI: http://doi.org/10.1111/fima.12300
• Author: Christine A. Parlour,Uday Rajan,Albert S. Hu
• Date: 01 December 2019
• Published date: 01 December 2019

DOI: 10.1111/fima.12300

ORIGINAL ARTICLE

Cryptocurrencies: Stylized facts on a new

investible instrument

Albert S. Hu1Christine A. Parlour1Uday Rajan2

1Haas School of Business, UC Berkeley,Berkeley,

California

2Stephen M. Ross School of Business, University

of Michigan, Ann Arbor, Michigan

Correspondence

ChristineA. Parlour, Haas School of Business, UC

Berkeley,Berkeley,CA 94720.

Email:parlour@haas.berkeley.edu

Abstract

We present stylized facts on the asset pricing properties of cryp-

tocurrencies: summary statistics on cryptocurrency return prop-

erties and measures of common variation for secondary market

returns on 222 digital coins. In our sample, secondary marketreturns

of all other currencies are strongly correlated with Bitcoin returns.

We also provide some investment characteristics of a sample of 64

initial coin offerings.

1INTRODUCTION

Cryptocurrencies are a new class of investibleinstruments and can evenbe included in individual retirement accounts.

As of November 2017, the total market capitalization of cryptocurrencies reached over $300 billion, with Bitcoin,

Ether, and Ripple being the most prominent. Briefly,cryptocurrencies or tokens are digital assets issued in return for

remuneration in the form of fiat money or other cryptocurrencies. Variousexchanges exist in which they can be traded

for other cryptocurrencies or fiat money. The convertibility of these digital assets to fiat money means theycan be

incorporated into any portfolio and viewed as any other asset class, even though their exact legal definition is still

unsettled.Although brokerage, trading, and even financial derivatives for these currencies are beginning to flourish and

receivewidespread attention, little research has been done on the asset pricing properties of this new instrument class.

In this paper, we document some stylized facts about cryptocurrency returns. Besides the novelty of this invest-

ment vehicle, cryptocurrencies provide an interesting benchmark because they are effectively unregulated. Specifi-

cally,prior to the SEC crowdfunding regulations, there was no restriction in the United States on who could introduce a

cryptocurrency.Further,trading is also unregulated, so prices must also reflect uncertainty associated with the viability

of an exchange.By contrast, most other data series examine assets that were issued in the context of some regulatory

framework; such data are therefore truncated by the minimum requirements of either exchangelisting, or adherence

to GAAP or other such requirements.

Both the returns and volatilities that we document are high. In addition, we show that cryptocurrencies, in aggre-

gate, carry a common source of systematic risk correlated with Bitcoin returns. This has important implications for

portfolio diversification and risk assessment. We also document the performance of initial coin offerings (ICOs) and

c

2019 Financial Management Association International

Financial Management. 2019;48:1049–1068. wileyonlinelibrary.com/journal/fima 1049

1050 HU ET AL.

compare them to Initial. Public Offerings (IPO)s. Interestingly, an economic implication of the ICO marketis that the

initial returns to IPOs may be puzzlingly low.

This paper contributes to an emerging literature on the pricing of Bitcoin, and the larger question of the economic

value of cryptocurrencies. A number of papers are concerned with explaining valuation and pricing of Bitcoin from

economicfirst principles. Athey,Parashkevov,Sarukkai, and Xia (2016) evaluate a model of adoption with Bitcoin prices

up to 2015 and concludes that adoption cannot explain prices (Yermack,2013). Ciaian, Rajcaniova, and Kancs (2016)

use an econometric approach to show macro-financials do not explain Bitcoin prices. Gandal and Halaburda (2014)

suggest a network effect is present that characterizes competition between different cryptocurrencies, and explain

Bitcoin’s early dominant position.

Inrelated work, Elendner, Trimborn, Ong, and Lee(2016) provide a brief history of altcoins and document secondary

marketreturn properties for Bitcoin and altcoins over the period April 2014 through July 2016. They also consider the

pairwise correlations of the top 10 cryptoccurrencies by market capitalization as of July 2016, and show the repre-

sentation of principal components in the returns of each of these currencies. They find that Bitcoin is not represented

in the first principal component, but is instead represented in the later ones. In contrast, we demonstrate that Bitcoin

returns have a high correlation with the first principal component of altcoin returns, so that Bitcoin may indeed be

thought of as a factor that drives cryptocurrency returns.

Various authors have investigated the properties of ICOs and cryptocurrencies, including Lyandres, Palazzo, and

Rabetti (2019), who reflect on the similarities between cryptocurrencies and other securities, Bourveau, DeGo-

erge, Ellahie, and Macciocci (2019), who investigate information provision on ICOs, while Lee, Li, and Shi (2019)

considers information aggregation. Finally, Howell, Niessner, and Yermack (2019) provide detailed analyses of ICO

markets.

A related strand of literature addresses market efficiency in cryptocurrencies, which is characterized by a high

degree of decentralization in trading and in issuances. Kroeger and Sarkar (2017), for example, show the law of one

price is often violated for Bitcoin, and relate this to the microstructure of Bitcoin trading. In a similar vein, Gandal,

Hamrick, Moor,and Oberman (2018) exhibit price manipulation in the Bitcoin market, likely by a single trader.We take

the prices (and hence the return patterns) of alternative cryptocurrencies (altcoins) as given, and document a number

of stylized facts.

The remainder of this article is organized as follows. Section 1.1 discusses the marketmechanisms of ICOs and trad-

ing. Section 2 gives an account of the data sources. Section 3 documents returns to ICOs and compares them to the

literature on IPOs. Section 4 discusses the return distributions of cryptocurrencies, and their correlation with other

assets. Section 5 discusses the possible benchmark to evaluate returns. Section 6 concludes.

1.1 Cryptocurrencies: Coins and tokens

Bitcoin is one of the earliest and well-known cryptocurrencies, and the first to use a blockchain to record and

decentralize the ledger of ownership and transactions and through it solve the “double-spend” problem. There are

two categories of tradable cryptocurrencies alternative to Bitcoin: coins and tokens. These are sometimes referred

to as altcoins (so called because they are coins that are alternative to Bitcoin), Although the nomenclature is not

standardized, there often are technological and use-differences between coins and tokens. Coins typically have their

own blockchains, whereas tokensare issued on some underlying platform, often one that enables smart contracts. For

example, the altcoin, Litecoin, is recorded on a variation of Bitcoin technology,others such as b-cash use new software

implemented as a fork on a pre-existing transaction ledger.1Coinsare used mainly as media of exchange or stores of

value, akin to nondigital currencies. Prominent coins other than Bitcoin include Ether and Ripple.2Tokensare typically

1TheBitcoin block chain, in this specific case.

2“Ether”is often used synonymously with “Ethereum” to refer to the currency, although the latter more correctly describes the entire platform.