Blockchain governance - Part 7
A few scenarios were considered for this analysis, being the most feared of them the ‘majority attack’, also called ‘51% hashpower attack’, when a powerful and wealthy adversary is able to acquire the same processing capacity, or decision power, as the honest nodes.
By Marcelo Martins - December 31, 2020
Part 7: Assessment results Permalink
Part 5 - Assessing blockchain governance and Part 6 - Testing governance hypotheses show details about how Decred improves security, keeps predictable coin issuance, maintains scarcity, balances power and incentives between PoW and PoS miners and runs its governance through transparent but private voting on on-chain and off-chain proposals.
A few scenarios were considered for this analysis, being the most feared of them the ‘majority attack’, also called ‘51% hashpower attack’, when a powerful and wealthy adversary is able to acquire the same processing capacity, or decision power, as the honest nodes. Although this scenario is extremely unrealistic for Decred, not only because it would consider an adversary buying 13 240 units of DCR5 ASIC devices or equivalent processing power almost at once, which implies that these devices must be already built and for sale (otherwise, as with any other coin, if the purchase is done during the course of a few months, the devices might become inefficient with the release of newer and more capable devices or the honest adversaries might increase their share) and mostly because considering the issued Decred coins until 10 Feb 2020, shown in Table 9 of Part 5 - Assessing blockchain governance, the adversary would have two options: i) an internal attack, which means buying half of the coins the stakeholders locked up in tickets (which would in turn require them to stop staking and selling them at market price) for the dishonest adversary to lock them back into tickets to reach 50% of PoS voting power; or ii) buy almost all of the remaining unlocked Decred coins, which by 10 Feb 2020 amounted to 50,16% of all issued coins (49,84% is locked up in tickets, so the adversary would have to buy another 49,84% to have the same voting power), which implies first in finding them because not all coins are available for sale in online exchange platforms and second, in buying them all without skyrocketing its price.
Simulation: the InvalidationGame Permalink
Part 5 - Assessing blockchain governance, subsection ‘Simulation: the InvalidationGame’ used a simulation script called InvalidationGame to demonstrate the cost of performing such an attack in terms of time and forgone block rewards (had the attacker chosen to perform an honest work), which also translates into wasted electricity costs. Tables 2 and 3 showed how long an attack takes in average when PoS is added as a second factor of authentication to PoW. While 100 000 45%/55% pure PoW simulated attacks took 35 minutes to complete, a batch of 100 000 45%/55% PoW + 70%/30% PoS simulated attacks took 4h03m to run its course until a 6-block difference between adversaries. For the previously mentioned ‘majority attack’ scenario, Tables 12 and 13 show that the same 100 000 simulations took 32 minutes and 7h02m, respectively. In terms of cost and forgone block rewards it is important to notice that the ‘attack cost’ shown on Tables 3 and 13 calculate only successfully PoW mined blocks invalidated by PoS miners, not the work done on PoW blocks wasted when the adversary was able to successfully mine them, as shown on Tables 2 and 12. So, although not shown, the attack cost on those PoW+PoS simulations is much higher than those numbers.
Security increases opportunity costs Permalink
Part 5 - Assessing blockchain governance, subsection ‘Security increases opportunity costs’ considered a mixture of an external PoW/internal PoS attack scenario to try to make it less unrealistic as previously explained, which means the attacker would have to buy ASIC devices on the market increasing the total number of PoW devices but the coins would have to be bought from current stakeholders willing to sell at market price, keeping the staked percentage at the same level.
Considering the information presented on Tables 12 and 13, it would be reasonable to assume that while a PoW attack would cost an average of 36 block rewards forgone because the dishonest adversary decided to use the processing power to try to attack the network as shown in Table 12, a PoW+PoS attack would cost the average of 18 block rewards from invalidated blocks due to PoS mining as shown on Table 13 + the 36 PoW block rewards from Table 12, because after mining a valid block through PoW mining, the adversary must have the block validated in PoS mining.
Tables 6 and 8 show the hourly cost to operate the ‘majority attack’ on Decred and Bitcoin, respectively. Because Decred network produces 12 blocks per hour and Bitcoin 6 blocks per hour, in average, and because of the number of block rewards forgone in an attack shown right above, an attack would take a few hours to complete, increasing its costs. An attack targeting Decred network would inflict an additional damage by leaving the dishonest adversary with coins locked up in tickets that shall (almost totally) lose their value when a successful (double-spending) attack becomes public.
Even extremely unrealistic because of the available Decred coins necessary to launch a PoS attack, Part 5 - Assessing blockchain governance, subsection ‘Security increases opportunity costs’ shows that an external/internal ‘majority attack’ on Decred is cheaper to attempt than on Bitcoin due to network size and USD exchange rates. Decred would be more expensive to attack if its USD exchange rate increased by 36 times and its network size increased by 100 times. Decred would also be 8,2753 times more expensive to attack if Decred and Bitcoin had the same network size and USD exchange rate. On the other hand, an internal/internal attack on Decred is much more expensive than an internal attack on Bitcoin blockchain.
Testing governance hypotheses Permalink
Part 6 - Testing governance hypotheses used real blockchain data to analyse Decred governance aspects and if price variance or staked percentage is somehow related to other factors, for example. When predictability and scarcity, introduced in Part 2 - The tech behind digital money are evaluated, observed Decred blockchain data analysed on hypotheses 5 and 6 show that coins are issued with the expected predictability and tickets are voted and rewarded as predicted in the official documentation, which is fundamental to produce scarce, valuable, trustworthy coins.
According to the data analysed on hypotheses 3 and 7, Decred USD exchange rates were correlated to Bitcoin USD exchange rates until mid-year 2018, when Decred ASIC devices were released. Somewhere between that point in time and Jan 2019 the data shows that Decred USD exchange rates became moderately correlated to Decred network difficulty (which increased after the release of ASIC devices as expected).
Hypothesis 8 considered the voting participation as a means to evaluate on-chain and off-chain governance. The data compiled and displayed on Figures 14 and 15 and Table 28 indicate that Decred stakeholders have more interest in on-chain proposals, those able to change blockchain consensus rules and cause long-term impact on the project. These proposals, written by core developers, may be more relevant and also well-written than average which is a possible explanation for the higher approval rate.
The staked percentage of Decred coins is related to the amount of issued coins but not to the USD exchange rate, according to hypotheses 1 and 2. Table 15 shows a strong correlation between staked percentage and issued coins, which indicates that while new coins were issued according to the rules, stakeholders showed an increasing interest in participating in the governance and in being passively rewarded in PoS staking. Regarding USD exchange rates, Table 16 shows no correlation with staked percentage one year after project launch, period selected to disregard the initial staking starting on zero as possible outliers.
OLS model and decision tree Permalink
The influence of average hashrate and staked percentage in price are depicted in the OLS model of Part 6 - Testing governance hypotheses, where the relation between the variables from 01 Jan 2019 to 10 Feb 2020 shows that a significant increase in average hashrate may double Decred price. The relation between average hashrate (PoW) and price indicates that PoW miners refrain from selling low and that PoW security has great influence in price, while the relation between demand for tickets (financial return + PoS voting power) and price possibly indicates that long-term investors are not willing to undo their positions in exchange for a small profit in the present while forgoing the opportunity to direct the project as they see fit and the risk of a much higher return on investment in the future. This relation between price, average hashrate and staked percentage is also shown in the regression tree, where the data shows that lower prices happened when the network had lower hashrates (PoW security).
Traditional economists are still sceptical regarding cryptocurrencies, attributing the value of money to men with guns (Krugman, 2018) or stressing the question of trust because “every currency is based on confidence that the hard-earned ‘deposited’ into it will be redeemable on demand” (Stiglitz, 2019), even though they are only redeemable in themselves and men with guns1 could not hold the purchasing power of the Venezuelan bolivar. They believe “conventional money generally does its job quite well”, that “using a bank account means trusting a bank, but by and large banks justify that trust” (Krugman, 2018) and that cryptocurrencies are used by “people who engage in nefarious activities” (Stiglitz, 2019).
This inquiry showed how secure blockchain technologies increase governance and avoid trusted third parties, providing security, transparency, incentives, change rules and network coordination. Real-world use cases, forgotten by the aforementioned Nobel laureates, include unbanked people in poor countries2, citizens of collapsing dictatorships3, remittances in developing countries4, refugees’ access to food and cash5, evasion of barriers to international payment systems and financial markets6, avoidance of censorship and state control7, avoidance of banks and third parties8 and freezing of assets9, increase in voting security, property transactions10, innovation hubs11, digital notaries, to fight Covid hardships12 and payments.
Even the large banks are creating their own private coins13, in a movement that resembles Hayek’s prediction previously explored in Part 2 - The tech behind digital money. The International Monetary Fund (IMF) is also considering the issuance of a digital currency, a centralised one, as expected (Lagarde, 2018).
Cryptocurrencies avoid the personal data leakages14 occurred to large banks, because they do not process personal data, only centralised exchange platforms do. Although banks claim they can refund customer’s stolen money, so far they weren’t able to ‘refund’ social security numbers and driver’s licenses as explained by Decred Lead Developer Marco Peereboom (Web Summit, 2019). To counter this issue, Decred developers released a free, open source, decentralised exchange platform called DEX15 to enable trustless exchange of blockchain assets through atomic swaps.
For the EU Blockchain Observatory and Forum, data privacy issues range “from reconciling blockchain’s data sharing properties with the data protection provisions of the GDPR to addressing the legal status of smart contracts and digital assets.” (European Commission, 2018). How would permissionless public blockchains comply with the right to erasure warranted in GDPR Article 1716? Or private or permissioned ones comply with the same right warranted by CCPA Section 1798.10517? Personal data should not be trusted in the hands of third parties and models have been proposed to reduce the trust and limit the access by data processors (Zyskind, Nathan, & Pentland, 2015).
The issue of fake news and disinformation in social media is not new, only the technology was upgraded. A NATO report on Russian gaslighting informs that after WWII “the USSR was not afraid to spread disinformation that clearly differed from reality; it used an integrated approach in distributing its propaganda through a variety of channels, ranging from educational institutions and scientific and reference literature to extensive use of the popular media” (Krumiņš, 2018). To fight Russian disinformation, Estonia created the Digital Embassy18 project using KSI blockchain19 technology. But fake news and disinformation come to one central issue: who has the power to assert the truth? The digital notary using blockchain, backed by good governance and strong security, may provide a solution to protect history. Solutions that use blockchain and smart contracts to check news (Qayyum, Qadir, Janjua, & Sher, 2019) and also that leverage Artificial Intelligence and Machine Learning to identify fake content (Hasan & Salah, 2019) have been proposed, but some of them end up being a high-tech version of the Ministry of Truth, depicted by George Orwell in his novel ‘1984’ from 1949. In other cases, such decentralised applications may be targeted by bots and fake accounts, the current problem with fake news but upgraded to the blockchain.
Decred Project shows one way of implementing blockchain governance and plays its role in several of the use cases mentioned in this section while adding a ‘second-factor of authentication’, simulated by the InvalidationGame, that depends on ‘skin in the game’. Skin in the game is key to a good governance. While some economists venture a critique, and at the same time, spare no effort to defend a State that operates “an economy that socialized risk and privatized rewards of economies in a manner that enriched elites at the expense of everyone else” (Mazzucato, 2013, p. 193), as mentioned in the Introduction, Decred Project operates a governance model where one has to be invested to make the calls in a direct democracy, after debating with peer investors or fellow citizens and then live with the consequences of their choices (Yocom-Piatt, 2019). After all, “if you have the rewards, you must also get some of the risks, not let others pay the price of your mistakes. If you inflict risk on others, and they are harmed, you need to pay some price for it” (Taleb, 2018, p. 11). Skin in the game explains why Decred Project funds itself and its innovations with the network fund, not with taxpayers’ money from an ‘Entrepreneurial State’.
Decred Project not only aims to provide an alternative to financial systems that depend on a trusted third party and to people that depend on the good will of giant financial institutions but also to show how e-voting may happen in a secure and transparent fashion. In 2018, the European Parliament mentioned e-voting as a possible use case in their Blockchain Resolution (European Parliament, 2018). E-voting systems need to be transparent, open source and extensively tested to provide a trustworthy digital identity system, allowing voters to be registered and cast their votes anonymously, and reveal an automatic count while preserving voters’ identity. The most recent attempt to succeed in building such voting system is the patent filed20 by the United States Postal Service (USPS) on Aug 13, 2020 after a few setbacks before the US presidential election. The patent describes a “Secure Voting System” that mixes blockchain and mail, where a registered voter receives a computer readable code by mail and confirms their identity. In the voting process, the system separates the voter’s identity and publishes the vote in the blockchain. Previously in April of the same year, the Utah Republican Convention successfully used blockchain for its election process21 with a large turnout. Almost a week after that, in Apr 30, 2020, US Senate staff suggested on a staff memorandum22 that the Senate could consider blockchain as a remote voting tool, citing the example of Estonia, if they could eliminate the threat of a ‘51 percent attack’. E-voting systems could help reduce power imbalances and information asymmetry, ensure that only eligible voters participate and increase voter’s turnout and the trust in elections and democracy.
1 “Venezuela’s Latest Problem Is There Are Now Too Many Dollars”, https://www.bloomberg.com/news/articles/2020-01-03/venezuela-s-latest-problem-is-there-are-now-too-many-dollars
2 “Can blockchain accelerate financial inclusion globally?”, https://www2.deloitte.com/content/dam/Deloitte/lu/Documents/technology/lu-blockchain-accelerate-financial-inclusion.pdf
3 “Bitcoin has saved my family”, https://www.nytimes.com/2019/02/23/opinion/sunday/venezuela-bitcoin-inflation-cryptocurrencies.html
4 “Moving beyond remittances to help the world’s poor”, https://www.weforum.org/agenda/2019/08/cross-border-money-for-the-poor
5 “WFP Building Blocks”, https://innovation.wfp.org/project/building-blocks
6 “Skirting U.S. sanctions, Cubans flock to cryptocurrency to shop online, send funds”: https://www.reuters.com/article/us-cuba-cryptocurrency/skirting-u-s-sanctions-cubans-flock-to-cryptocurrency-to-shop-online-send-funds-idUSKCN1VX2H0
7 “Why China Is So Afraid Of Cryptocurrencies”, https://www.forbes.com/sites/rogerhuang/2020/12/31/why-china-is-so-afraid-of-cryptocurrencies/
8 “Why are Venezuelans seeking refuge in crypto-currencies?”, https://www.bbc.com/news/business-47553048
9 “Brazil tries shock therapy on inflation”, https://www.chicagotribune.com/news/ct-xpm-1990-03-25-9001240707-story.html
10 “Chromaway property transactions – Sweden”, https://publications.jrc.ec.europa.eu/repository/bitstream/JRC115049/blockchain_for_digital_government_online.pdf
11 “The World Bank is betting big on blockchain-based bonds”, https://www.technologyreview.com/2018/08/10/141094/the-world-bank-is-betting-big-on-blockchain-based-bonds/
12 “UNICEF Cryptocurrency Fund”: https://www.unicef.org/press-releases/unicef-cryptocurrency-fund-announces-its-largest-investment-startups-developing-and
13 “The end of Ripple?”, https://www.forbes.com/sites/billybambrough/2019/02/15/the-end-of-ripple/
14 “More Than 100 Million Consumer Personal Data Leaked After A Massive Cloud Breach At Capital One”: https://www.forbes.com/sites/jeanbaptiste/2019/07/30/more-than-100-million-consumer-personal-data-leaked-after-a-massive-cloud-breach-at-capital-one/
15 Source code is available at https://github.com/decred/dcrdex
16 EU General Data Protection Regulation (GDPR) is available at https://eur-lex.europa.eu/eli/reg/2016/679/oj
17 CCPA available at https://leginfo.legislature.ca.gov/faces/billTextClient.xhtml?bill_id=201720180SB1121
18 Available at https://e-estonia.com/estonia-to-open-the-worlds-first-data-embassy-in-luxembourg/
19 Available at https://e-estonia.com/solutions/e-governance/data-embassy/
20 US Patent Application Publication no. US 2020/0258338 A1, Secure Voting System, available at https://pdfaiw.uspto.gov/.aiw?docid=20200258338&SectionNum=1&IDKey=7A4F4EA40D1F&HomeUrl=http://appft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1%2526Sect2=HITOFF%2526d=PG01%2526p=1%2526u=/netahtml/PTO/srchnum.html%2526r=1%2526f=G%2526l=50%2526s1=20200258338.PGNR.%2526OS=%2526RS=
21 Utah Republican Convention Uses Blockchain Voting Service, available at https://www.forbes.com/sites/robertanzalone/2020/04/29/utah-republican-convention-uses-blockchain-voting-service/#39d925edcfee
22 PSI Staff Memorandum, available at https://www.portman.senate.gov/sites/default/files/2020-04/2020.04.30%20PSI%20Staff%20Memorandum%20-%20Remote%20Senate%20Process%20-%20Final_0.pdf
European Parliament. (2018, October 03). Distributed ledger technologies and blockchains: building trust with disintermediation. Retrieved from European Parliament: https://www.europarl.europa.eu/doceo/document/TA-8-2018-0373_EN.html
Hasan, H. R., & Salah, K. (2019). Combating Deepfake Videos Using Blockchain and Smart Contracts. IEEE Access, 7, pp. 41596-41606.
Krugman, P. (2018, July 31). Transaction Costs and Tethers: Why I’m a Crypto Skeptic. Retrieved from The New York Times: https://www.nytimes.com/2018/07/31/opinion/transaction-costs-and-tethers-why-im-a-crypto-skeptic.html
Krumiņš, G. (2018). Soviet Economic Gaslighting of Latvia and the Baltic States. Defence Strategic Communications, 4(Spring 2018), pp. 49-78. Retrieved from: https://www.stratcomcoe.org/gatis-krumins-soviet-economic-gaslighting-latvia-and-baltic-states
Lagarde, C. (2018, November 14). Winds of Change: The Case for New Digital Currency. Retrieved from International Monetary Fund: https://www.imf.org/en/News/Articles/2018/11/13/sp111418-winds-of-change-the-case-for-new-digital-currency
Mazzucato, M. (2013). The Entrepreneurial State. Anthem Press.
Qayyum, A., Qadir, J., Janjua, M. U., & Sher, F. (2019, July-August). Using Blockchain to Rein in the New Post-Truth World and Check the Spread of Fake News. IT Professional, 21(4), pp. 16-24.
Stiglitz, J. (2019, July 02). Thumbs Down to Facebook’s Cryptocurrency. Retrieved from Project Syndicate: https://www8.gsb.columbia.edu/articles/chazen-global-insights/thumbs-down-facebook-s-cryptocurrency
Taleb, N. (2018). Skin in the Game: Hidden Asymmetries in Daily Life. Random House.