In our paper, "ICO investors" (available on SSRN), we investigate the composition and trading behavior of the investor base in initial coin offerings (ICOs). ICOs have become the prevalent source of financing for start-up companies that use blockchain technology, raising more than 30 billion dollars in funding to date. However, so far, relatively little is known about their investors.
We study a hand-collected sample of over 300 successful ICOs that ended before April 2018 and each raised at least 1 million dollars in funding. In ICOs, investors receive cryptographic tokens in exchange for their investment. Most often, these tokens are hosted on the Ethereum platform. We study public transaction data from Ethereum to determine the composition and secondary market trading behavior of the investor base. We find that the average ICO has 4,700 contributors, and that the median contribution to the average ICO is only $1,200. Based on these results, the representative participant is likely a retail investor. We also find that almost half of all contributors sell some or all of their tokens in the secondary market within ninety days of the ICO, suggesting that most contributors are motivated by financial gains.
A challenge when analyzing Ethereum data is that they are pseudonymous, i.e. transactions are only identified by an Ethereum address (the equivalent of a bank account number) while the ultimate beneficial owner behind the address remains anonymous. For our analysis, we equate wallets with individuals and tokens moving between wallets as trades. We provide evidence suggesting that many investors indeed only use a single Ethereum address to participate in an ICO, and that therefore this approximation is reasonable. First, we find that the aggregate daily volume of tokens being moved out of ICO investors’ wallets is highly correlated with the daily trading volume on cryptocurrency exchanges. This result indicates that investors usually move tokens to sell them rather than to move them to another one of their wallets. Second, following the ICO, participating wallets are frequently used for sending and receiving Ether, the Ethereum network’s own cryptocurrency, which suggests that they are generally not special purpose wallets created for a single investment. Finally, the number of participants publicly disclosed by ICOs with a KYC procedure (where the issuer knows the number of individual participants) is statistically indistinguishable from the number of participating wallets.
ICOs are typically conducted in two stages: a closed presale stage for large investors, often taking place at discounted prices, and a public crowdsale stage. If the prevailing secondary market price after the ICO is at or above the presale price, which is lower than the “list price” paid by regular investors, presale investors can realize a profit by selling immediately. We find evidence that they do. Large investors sell earlier if there was a presale and if the presale discount was high, and holding period returns to other investors are decreasing in the amount of funding raised in the presale as well as the presale discount.
Most companies financing themselves through ICOs have unproven business models and are in the pre-product stage. There exists virtually no hard information on them, so asymmetric information is large. The financing of such early stage companies has previously been the domain of specialized angel investors or venture capitalists who acquire soft information by meeting with potential customers, suppliers, and the founding team, and who demand securities guaranteeing priority and control rights. We investigate whether the ICO market has emulated some of these investor protections. We find that ICO investors receive very few rights compared to traditional early stage investors. The majority of ICO tokens are utility tokens that can be redeemed in exchange for a product or service in the future, but do not confer any cash flow rights. Less than a fifth of ICOs give participants voting rights, and such votes are often only advisory in nature. None of our sample ICOs allows investors to participate in director elections and none gives them liquidation preferences. The only provision from venture capital contracts that is also common in ICOs are vesting periods. A majority of ICOs lock up tokens held by founders and the issuer, for an average period of 1.1 years.
Rüdiger Fahlenbrach, Ecole Polytechnique Fédérale de Lausanne (EPFL)
Marc Frattaroli, Ecole Polytechnique Fédérale de Lausanne (EPFL)