Project Blog: How does the Cryptocurrency market perceive regulation?

Should the cryptocurrency market be regulated?

In short:

  • Despite calls for, and ad hoc attempts of, regulation in the cryptocurrency market, the benefits and costs of regulation are unclear. We attempt to provide empirical evidence for this issue by investigating what the market’s perception of regulation is.

  • We find that cryptocurrency investors on average perceive the increased (decreased) likelihood of regulation to be net costly (beneficial).

  • However, specific characteristics relating to ‘quality’ and ‘transparency’ attenuate this reaction: market reactions are less negative when (1) experts ratings about transparency, management and vision are higher, (2) cryptocurrencies are more interactive on twitter.

  • Regulation is less costly for the cryptocurrencies that are expected to be more capable at handling the potential burden of regulation changes, or the cryptocurrencies that have a better information environment.

While the financial markets authorities in most countries took a hands-off approach for a long time, pressure is now building on regulators to take action in the cryptocurrency market. However, it is not clear whether regulation is net beneficial or costly for investors in the cryptocurrency market. Primarily because cryptocurrencies — due to their nature — are difficult to reconcile with existing regulatory frameworks.

source: Cointelegraph

source: Cointelegraph

A lot of the recent debate on regulation and cryptocurrencies is focused on how the frameworks can and ought to be applied from the perspective of the regulatory agencies. However this project, together with my PhD supervisor Edith Leung, focuses instead on the investors’ perception of regulation: by examining how the market reacts to news about regulation we can gauge if the cryptocurrency investors themselves view regulation as something beneficial or costly. This setup furthermore allows us to investigate what and how the differences in cryptocurrency characteristics affects the market reactions to regulation news, by focusing on a subset of cryptocurrencies called ‘tokens’.

We use a number of web-crawlers in R to gather the data needed for this study. Specifically, we scrape news items from Cointelegraph (https://cointelegraph.com/), cryptocurrency market data from Coinmarketcap (https://www.coinmarketcap.com), ICO data from ICOBench (http://www.icobench.com), and Twitter data from Twitter (https://twitter.com/).

The first difficult part of the analysis is determining what the news is to which we expect the market to react. We gather and manually classify a total of 1,009 news items from the period Sept. 2013 — Aug. 2018 of which we expect 148 events to have news value and also relate to a country with a top 100 cryptocurrency exchange. An example of a news event that increases the likelihood of regulation is the following: “South Korea Moves to Impose Stricter Digital Currency Regulations” (Sept. 5th, 2017). An example of a news event that decreases the likelihood of regulation is: “Singapore Will Not Regulate Cryptocurrencies, Singapore Official Says” (Oct. 26th, 2017).

Another difficult consideration is to determine what the ‘market reaction’ actually is. A research design in which you investigate market reactions to news is called an ‘event study’. As is usual for an ‘event study’, we try to measure the abnormal market reaction, i.e. the reaction of the market on the date of the event that goes beyond what is normally expected:

abnormal market reaction = market reaction on event date — average market reaction

Practically, this means that we subtract the average daily return in the period surrounding the event, from the return on the event date. The sum of this abnormal market reactions on the event date and the date after the event date is called the cumulative abnormal return, the main variable of interest in this study.

Average market reaction

Below, I present the average cryptocurrency market reactions to the news events we identified. As can be seen, the market reacts on average negatively to news events that increase the likelihood of regulation. If regulation is perceived as net costly by market participants, they are expected to react positively to news that decreases the likelihood of regulation. As can be seen in the last column of table 1, we find this to be the case.

Table 1:  Cryptocurrency market reaction to regulation news. ‘Increasing regulation’ (‘Decreasing regulation’) are those news events that are expected to increase (decrease) the likelihood of regulation.

Table 1: Cryptocurrency market reaction to regulation news. ‘Increasing regulation’ (‘Decreasing regulation’) are those news events that are expected to increase (decrease) the likelihood of regulation.

What characteristics affect the market reaction?

Next, we want to investigate what and how cryptocurrency characteristics affect the average market reaction to regulation news. For this analysis, we focus on a subset of cryptocurrencies that are issued through a process called an ‘ICO’ (Initial Coin Offering), in which a company sells their cryptocurrencies to raise funding for starting- or scaling-up.

Investors might view regulation to be costly, but might also think that cryptocurrencies with a more experienced management team to be more efficient or competent in dealing with new regulatory requirements. Also, proposed regulations that would require more transparent disclosure are likely to have less effect on cryptocurrencies that already disclose more information. Finally, if stricter regulation for exchanges leads to ‘higher quality’ cryptocurrencies to be listed on average, we expect less impact of such regulations on those higher quality cryptocurrencies.

In short, we would expect the overall negative reaction to regulatory news to be less pronounced for more transparent or ‘higher quality’ cryptocurrencies. We use ratings data from ICOBench to gauge the quality of cryptocurrencies and cryptocurrency Twitter activity to capture their transparency. We estimate this using ordinary least squares regressions (OLS).

ICOBench is a platform that facilitates the investments in ICOs by providing overviews of ongoing and past ICOs. Interestingly, ICOBench beside ‘rating’ the information provide by the ICO’ing company algorithmically in what is called the ‘Benchy Rating’, ICOBench also has an active panel of experts that rate the issuing company’s management, vision, product idea. Together, these two ratings are combined in the ‘Total Rating’.

Table 2 shows that the ‘total rating’ is not significantly associated with the market reaction to regulation news. ‘Benchy rating’, which is an algorithmic assessment of all available information of the ICO and is comprises 4 subratings, is also not significantly associated with the market reaction. Importantly, the ‘social media score’, which measures the presence and activity of the cryptocurrency on social media, is positively associated with the market reaction. This means that the more active the cryptocurrency is on social media, the less negative the effect to regulation news.

In addition, the ‘Expert Rating’ and all subcomponents are all positively significantly associated with the market reaction. This means that the market reaction to regulation news is less negative for cryptocurrencies of which the ICO is higher rated by experts. Therefore, the experts opinion signals something that is valuable to cryptocurrency investors for their assessment of the effect of regulation news.

Table 2:  the effect of different cryptocurrency characteristics on the market reaction (CAR, cumulative abnormal return) of the news event.

Table 2: the effect of different cryptocurrency characteristics on the market reaction (CAR, cumulative abnormal return) of the news event.

If we then look at the differences per cryptocurrency in disclosure through Twitter in Table 3, several things stand out. First, the total number of Tweets of the cryptocurrency company do not affect the market reaction to regulation news. Secondly, the total favorites on and retweets of the Tweets of the cryptocurrency company also are not related to the market reaction. What does matter is how interactive the cryptocurrency is on Twitter: the more the cryptocurrency responds to questions of followers and clarifies their own posts, the less negative the market reaction to regulation news.

Table 3:  the effect of Twitter activity on the market reaction (CAR, cumulative abnormal return) to the news event.

Table 3: the effect of Twitter activity on the market reaction (CAR, cumulative abnormal return) to the news event.

Conclusion

Together, the results suggest that on average the market reacts negatively (postivively) to the increased (decreased) likelihood of regulation of cryptocurrency markets. This reaction differs from cryptocurrency to cryptocurrency, since we document that it is less negative for cryptocurrencies that are higher rated on management competence, product idea and transparency. In addition, cryptocurrencies that interact more with followers also have less negative reactions to news about increasing regulation, suggesting the value of a transparent information environment.

We conclude that the costs of regulation are lower for cryptocurrencies that are deemed to be more capable at handling the burden of regulation changes, or those that are less affected by regulation because they are already more transparent.

In the standard, heavily regulated financial markets it is difficult to assess the benefit of regulation for investors. The cryptocurrency market, as a sizeable market without a regulatory framework, offers to be a setting to investigate what the effects of introducing regulation is and what company characteristics make regulation less costly to investors.

Want to know more?

We are presenting this work at the European Accounting Association Annual Congress 2019 in Paphos, Cyprus. We have also posted the full paper online on SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3339197.