Do national development factors affect cryptocurrency adoption?

The adoption of cryptocurrencies is uneven across businesses, industries, and countries. Different forces drive cryptocurrency adoption (CA) dependent on the national level of development. We empirically assess the relationship between certain macro-national developmental indicators and cryptocurrency deployment across 137 countries. Linear regressions determine specific associations with cryptocurrency adoption. We report that CA correlates positively and in decreasing order with Education, the Human Development Index, the Network Readiness Index, the Gini index, Democracy, Regulatory Quality, and Gross Domestic Product, and negatively and in decreasing order with Control of Corruption, the Corruption Perception Index, and the Economic Freedom Index. We draw on our findings to point to policy implications tied to the usage of cryptocurrencies and blockchain technologies more widely and identify further research possibilities.


Introduction
Technological innovations impact organizations and countries differently (Wang et al., 2020).Cryptocurrency blockchain technologies adoptions are no different.But research in this area of innovation has tended to be prescriptive and theoretical focusing primarily on blockchain and cryptocurrency technical attributes with little scholarship undertaken on what drives differences across nations (Bhimani et al., 2021;Kouhizadeh et al., 2021;Schlecht et al., 2021).It is known that many developmental issues influence the adoption of cryptocurrency and blockchain systems in both developing and advanced economies (Behnke and Janssen, 2020;Bodkhe et al., 2020;Laroiya et al., 2020;Sinha et al., 2020;Stockburger et al., 2021).In relation to nations being receptive to cryptocurrency deployment, the focus has been on the potential costs and benefits offered by permissionless systems in different settings.But as Saiedi et al. (2021, p. 354) note: "While theoretical papers are emerging, discussing why cryptocurrencies, or digital currencies in general, may be adopted by individuals or businesses, there is a scarcity of global empirical studies on drivers of their adoption."Our aim is to point to macro-national developmental indicators that influence cryptocurrency adoption (CA) across economies.
We recognise there are different theoretical foundations as to the process of technology adoption and its role in economic development.
Theorists have advanced different perspectives including the work of Dixit and Pindyck (1994), Mansfield (1993), and Tornatzky et al. (1990) and others (Fan et al., 2018).We draw more explicitly on the economic growth and long run economic development arguments posited by Acemoglu (2009).Our interest lies in development factors affecting CA across nations.In reporting on factors which find positive or negative association with national developmental indicators, we contribute to policy decision making by identifying the strength of impact of different developmental indicators which are context-specific within different nations.We draw on data relating to 137 countries to ensure the comprehensiveness of our findings which have relevance to policy decision making.Aside from the practical implications of our findings, we advance academic research by widening the diversity of factors, both economic and societal, which impact CA.Scholars will benefit from our results in being able to extend the baseline research we have produced to other regional and economic contexts.The methodology we adopt will also prove useful in analyses tied to cryptocurrency and blockchain systems usage in other domains and point to further additional relationships with other independent variables and dependent variables.The rest of this paper is structured as follows.In Section 2 we identify some costs and benefits of cryptocurrency adoption as indicated by differing country deployment experiences.In Section 3, we discuss national development factors affecting cryptocurrency technology adoption decisions in developed and developing economies.In Section 4, we model cryptocurrency adoption and empirically assess the model in Section 5. We thereafter discuss and present limitations and identify areas for future research in Section 6. Section 7 concludes the paper.

Cryptocurrency adoption and experiences
Much has been documented on the ascent of the fourth industrial revolution.Artificial intelligence, information communication and technology, the internet of things and blockchains rest at the core of this fourth revolution.In respect to the latter, PwC (2020) predicts that blockchains will boost global GDP by US$1.76 trillion by 2030.Asia is expected to benefit the most from blockchains.Chinese bureaucrats place blockchains as one of the country's their top five priority expecting the potential net benefit in China to be US$440bn (versus US$407bn for the USA).Other countries including Germany, Japan, the UK, India, and France will see estimated returns exceeding US$50bn (PwC, 2020).At the enterprise level, across several developed nations, a majority of senior industrialists seem to be making blockchain systems a top priority for their organizations (Deloitte, 2020).
A major UNCTAD ( 2021) study (2021, p. 6) on blockchain usage notes that governments of developing countries "…should seek to strengthen their innovation systems to strategically position themselves to benefit from this new wave of technological change."In this light, Domjan et al., (2021, p. xi)) states that: "There is a growing realization all over the world, but especially in developing countries, that there is a set of problems linked to trust, verification and value transfer that could be solved with blockchain technology".Cryptocurrency adoption (CA) resting on blockchain logic can enable readier access to digital financial products and offer services at a lower cost increasing financial inclusion and connecting local populations to broader global markets (Aysan, 2020;Schuetz and Venkatesh, 2020).In developed nations, enterprises have been open to CA.For instance, firms such Tesla has accepted Bitcoin on-and-off for car payments (Hussain and Balu, 2021) and all goods and services can in principle be priced in Bitcoin (Frankenfield, 2021).Institutionally, there is also growing interest in cryptocurrencies for speculative, investment, store of value, and payment purposes.An International Finance Corporation study (IFC, 2017) reflected on how CA can promote "… greater financial inclusion and improve productivity" in developing economies.
Other economic benefits of CA viewed as relevant to developing countries include the protection of ownership rights, verifiable identity systems and the containment of corruption potential (Pisa and Juden, 2017).But the lack of standardization of transactional systems, weakness of governance institutions, high upfront installation costs, systems expertise and energy consumption issues are challenges affecting CA (Chang et al., 2020;Uddin et al., 2021).Bitcoin represents the first mobilisation of blockchain mechanics.Bitcoin is premised on the notion of a blockchain being a distributed ledger system which produces secure tamper-proof records.The elimination of third parties, immutability, transparency and decentralization enable a wide variety of applications for blockchain based systems of control (Iansiti and Lakhani, 2017).The high levels of privacy and anonymity allows for contracts execution more economically, rapidly and efficiently relative to contemporary systems relying on mechanical or human input (Angelis and Da Silva, 2019).But there are hurdles including high installation costs, operational skills requirements, systems changeover hurdles and regulatory constraints.A key point of significance is that blockchains may be permissioned or permissionless.Permissioned blockchains are used by centralized authorities in various forms (Allen et al., 2020;Kiff et al., 2020) whereas permissionless systems, with their protocol designs and processes of validation and mining transactions, ordinarily require crypto asset issuance.Nakamoto (2008) described a permissionless blockchain as exemplified by Bitcoin.While technical factors pervade cryptocurrency operations, CA entails dimensions of industrial capacity, regulatory environment, expertise availability as well as social and cultural influences (Hughes et al., 2019).Cryptocurrencies use specifically and blockchain systems deployment generally can be complex and entail a variety of issues that are tied to macro-level national factors.This represents the focus of our interest in this paper.We elaborate on this in the next section.

Developmental factors affecting cryptocurrency adoption
The question of whether information technologies generally show differential penetration and speed trajectories across developing and developed economies remains open due to absorptive capacity being influenced by a wide range of factors tied to level of national development (Keller, 2004;Niebel, 2018;Steinmueller, 2001).In this respect, we look at macro-level developmental indicators which impact technology adoption of which cryptocurrency deployment is a case.The literature on digital technology adoption at the country level is vast.Reference has been made to leapfrog possibilities, whereby processes in the accumulation of fixed investments and human capabilities can be sidestepped permitting developing nations to evolve faster as they can bypass having to disinvest prior structural institutions.Additionally, developing nations become more and more disconnected and less dependent on advanced country technical systems, have access to more knowledge and benefit as the costs of newer technologies decrease (Tchamyou et al., 2019).Venkatesh et al. (2003) brings together much of this literature in terms of perceptions of adopters and technology acceptance.Datta (2011) compliments this perspective with Rogers (1983) technology diffusion theory.In relation to macro level national contingent factors which enable or obstruct digital technology adoption at the national level, the prior literature points to technology support mechanisms (Kirkman et al., 2002) which include four distinct elements: economic context, policy and the legislative environment, society with a focus on education and training and access as per infrastructure capabilities.Economic conditions relate to the degree to which a technology may become integrated into the economic activity of a country.Policy focused issues relate to the overall business and economic climate of a country taking account of the level of competition permitted in terms of start-ups, technology development subsidies, privatization, and national legislative policies.Society-based elements include education and training.Access entails telecommunications and information infrastructure availability.Fong (2009) extends the literature analysis of factors relevant to conditions that may enable "technological leapfrogging" to include human capabilities, governance controls, institutional capacity and readiness, inequality indicators.The prior literature combined with Sen (2001) identification of democracy and human development in enabling the national development of technological advances leads us to focus on the following factors as impacting cryptocurrency adoption decisions: the legal environment, governance structure, democracy variables, human development, GDP, income inequality, education, economic freedom, and network readiness.We elaborate below on each of these and their relevance to cryptocurrency adoption decisions.

The legal environment
Not only has the legality status of cryptocurrency been under scrutiny but it has been long accepted that cryptocurrency would only survive and spur in countries with receptive regulatory environments.While the future of cryptocurrency is uncertain in most countries, other applications of blockchain including identity management, execution of smart contracts, supply chain routing and integrity are seeing growing usage.The legality status of blockchain is currently being debated at national levels primarily in reference to its use in financial and investment sectors.While a few countries have introduced regulations to support the implementation and use of the cryptocurrency, some countries have outrightly banned it and others have restricted its use in banking.Wright and De Filippi (2015) recognize the development of a new body of law to administer self-executed smart contracts and decentralized organizational structure while controlling for illicit activities.Cryptocurrencies have also been seen as enabling the replacement of tax havens (Marian, 2013), usage for money laundering (Barone and Masciandaro, 2019) and been associated with anonymity that encourages their use in digital black markets.Considering the challenge of taxing and regulating cryptocurrencies some of these exchanges have relocated to countries with friendlier or non-existent asset investment legal structures (Molloy, 2019).Akins et al. (2014) speak to the lack of federal income tax laws on virtual economy elements and propose a range of solutions.Switzerland, the UK, Estonia, Gibraltar and Malta have been identified as leaders towards the development of friendly yet secure regulation system (Suciu et al., 2019).Blockchain Alliance founded in 2015 has been actively working to conduct educational programs for around 700 law enforcement officers and regulators from more than 35 countries and include over 100 blockchain companies along with regulatory agencies including US, Europol and Interpol and authorities in Europe, Latin America, Africa, Asia, and Australia (Dewey, 2019).India plans to introduce laws to limit private cryptocurrencies and introduce a framework for the creation of an official central bank digital currency.Turkey has banned the use of crypto assets in payment services (Finextra, 2021) just as China continues to carry nationwide crackdowns against cryptocurrency mining shutting down more than 90% of its bitcoin mining capacity (Global Times, 2021).The ban on cryptocurrency however does not negate the benefit of the underlying technology via other applications.While India is exploring its potential benefits in smart contracts, the Blockchain Service Network (BSN) launched by China in 2020 seeks to enable enterprises to access, build and adopt blockchain for economic growth (Ozden, 2021).

Governance standards
The use of E-governance has long been debated to improve the governance systems through increased transparency, removing information asymmetry, minimizing delays and protecting against data theft (Banerjee et al., 2020;Halachmi and Greiling, 2013).The benefit offered by cryptocurrency including its temper free record keeping, decentralization and elimination of intermediaries makes it a cutting-edge anti-corruption technology for governance systems.Research is being conducted to assess these benefits of cryptocurrencies as well as central bank digital currencies to deal with issues of funds embezzlement and illicit activities especially in developing countries (Sanka and Cheung, 2019;Zbinden and Kondova, 2019).Government efficiency has been known to hasten technology adoption by lowering production costs, uncertainty imposed by corruption and protection of property rights (Galang, 2012;Murphy et al., 1991).Luo (2005) investigates the effects of corruption on innovative activity from the perspective of organizational theory.In this regard a firm will innovate or adopt new technology depending on the ease of the alternatives available.In the case of cryptocurrency, since it provides an audit trail and makes counterfeiting almost impossible, corrupt intermediaries may see it as a threat and discourage CA.
Cryptocurrency may be adopted as a solution to the corruption emanating from poor governance which has been a major cause of poverty in developing economies.Resnick (2020) studies the tax compliance behaviour of informal workers and concludes that the compliance is higher when the route of accountability between tax collector and payer is shorter.This can be achieved using cryptocurrency which eliminates intermediaries and offers higher level of trust in the system.On the other hand it has been argued that in well governed states the cost of replacing already established systems can exceed any benefits realized and so poorly governed states may make a fast paced shift to adopt cryptocurrency to address issues of legitimate transactions and to enhance transparency (Chan et al., 2008).Reducing corruption may be a priority for some emerging economies to facilitate global exchanges, economic development, and aid receipt.

Democracy level
Democracy is seen to reduce digital divide by promoting innovation and rapid technology adoption (Gao et al., 2017;Milner, 2006).Blockchain based solutions confer more power to citizens by reducing information asymmetries via decentralized voting, identity management, e-governance and e-democracy processes.Aysan et al. (2019) find that price volatility and returns of Bitcoin are positively and negatively, respectively, related to geopolitical risks.They suggest that Bitcoin can be a hedging tool against geopolitical risks.Moreover, since cryptocurrencies cannot be counterfeited, they are the perfect solution to the problems of decentralization and poor governance in many democracies.Tellman et al. (2021) highlight how decentralization of land regulation has left the system fractured.The informal urbanization of Mexico City at the hands of intermediaries and politicians had led to the exploitation of informal settlers.Boret et al. (2021) also confirm that decentralization of local expenditure could help lower poverty.But decentralization requires either trust in the local government or robust system which makes misappropriation impossible.Cryptocurrencies address these problems and so it is not only more likely to be adopted by democratic countries but is also further strengthen them.

Human development
The human development index is a composite measure of life expectancy, education and per capita income.Asongu and Le Roux (2017) show how information communication and technology could be used for human development in sub-Saharan Africa.Technological advancements have been shown to play a role in gender empowerment (Crittenden et al., 2019), providing access to healthcare (McGhin et al., 2019) and identity management (Kuperberg, 2019).Healthcare can realize the benefits of blockchain in various regards.It can revolutionize drug discovery and development process with better data availability, ability to carry out analytical procedures, optimizing the efficiency of the Internet of Healthy Things, safeguarding internet-connected medical equipment and combating counterfeit medicines (Clauson et al., 2018).Pharmaceutical companies on average lose $200 billion to counterfeit drugs annually.The World Health Organization has identified the worth of black market of medicine at $75 billion.Blockchain enabled traceability, auditability and secure databases for storing and accessing drug trial data can assist in reducing falsified medication extensively (Arsene, 2019).The storing of a patient's data on blockchain network giving him or her the right to share it with can enable faster sharing of information, help eliminate data corruption and the illegal selling of data (Hughes et al., 2018) Blockchain enabled identification can lead to certain benefits tied to human development.The World Food Program for instance, successfully deployed blockchain for its refugee program in Jordan.Various other countries are using blockchains to facilitate human development including Ghana which is using blockchain to resolve land disputes and India where the government think tank is working to use the technology in land titling and electronic health records (Nayyar, 2018).Smith and Floro (2020) highlight the significance of migration and remittance flows in reducing food insecurity especially in low-and middle-income countries.Such economies can multiply these benefits using blockchain.Apart from ensuring traceability, transparency, reducing cost through elimination of intermediaries, blockchain based solutions including cryptocurrency also has the potential to minimize bureaucracy and promote coordination amongst donors to ensure the best deployment of resources (Galen et al., 2018;Pisa and Juden, 2017).Cryptocurrencies also boost economic activity by including the unbanked into the financial system and empowering entrepreneurs to receive payments in more currencies.

GDP level
According to the growth principle technological transformation leads to an increase in GDP through an increase in capita per person which motivates savings and investment.Comin and Hobijn (2003) further support this claim by showing that the rate of adoption of technology in early stages of life cycle is largely determined by level of economic development of a country.While technology is a major determinant of economic growth a country's GDP per capita would influence technology adoption.Developing countries while most of them are highly indebted, face high inflation and low GDP would prioritize basic needs of the public over technology adoption owed to the high upfront implementation cost and investment in research and development of blockchain based application including cryptocurrency.Since cryptocurrency skills are still a niche market, the developers and network engineers are costly in terms of salaries (Davies, 2019).Organizations can be faced with the burden of hiring staff including compliance and legal personnel who understand the technology and can work in coordination with system developers and financial regulators.Another cost of implementation is that of energy consumption.Proof of work cryptocurrency require huge amount of energy, the Digiconomist's Bitcoin Energy Consumption Index estimated that one Bitcoin transaction takes 1544 kWh to complete (Gonzalez, 2021).This poses a hurdle for CA particularly in developing countries with high-cost energy sources.

Income inequality
Technology adoption in emerging economies has been slower as compared to developed economies given the cost of adoption and complementary factors like human capital (Comin and Mestieri, 2018;De Gregorio, 2018).The technology can create jobs for technical people while eliminating certain jobs worsening social and income inequalities.Analysing the opposite causation of that assessed in the current article, technological change can also reduce income inequality (Adrián et al., 2019;Tchamyou et al., 2018) in that many start-up companies like are able to be launched (Kshetri, 2017).Income inequality can have a cause-and-effect relation on cryptocurrency and ultimately the adoption of blockchain.It can motivate countries with high inequality to adopt cryptocurrencies to address the prevailing level of inequality.With lower transaction costs, financial inclusion and property rights security, and various income inequality issues, cryptocurrency adoption is regarded in some nations enable increased financial participation by the poor.Cryptocurrencies may confer greater economic participation since anyone with a smart phone and internet can become part of the global economy.Blockchain micro lending apps deployed in Southeast Asia have enabled 1.7 billion unbanked people in the world to build an auditable credit history.A digital currency created by the Venezuelan government has been deployed to assist citizens to be shielded from fiat currency devaluation (Carter, 2020).

Education level
Countries ranking higher in education level attainment can be expected to adopt cryptocurrency faster in the light of technical skills and knowledge presence (Li et al., 2019;Riddell and Song, 2017).Equally, the deployment impact of cryptocurrency can be extremely high in developing nations depending on skills level (Shapiro and Mandelman, 2021).Aside from pedagogical implications, blockchain systems can allow the issuance of decentralized degree certificates and checks on the authenticity of existing degree awards (Raimundo and Rosário, 2021).
Cryptocurrency usage can open up possibilities for smart contracts to be used as financial literacy and educational attainment can enhance the understanding of cryptocurrency based systems including the wider deployment of blockchains.

Economic freedom
Economic freedom includes the right to control one's labour and property.Since economic freedom is a measure of regulatory efficiency, financial freedom, and rule of law amongst other factors it could be closely linked to cryptocurrency adoption the success of which also relies on the aforementioned factors.Cryptocurrencies can help advance the core tenets of economic freedom including property rights, lack of reliance on central authorities, privacy and equality of opportunity (Gulker, 2017).They can eliminate the need of intermediaries and provide opportunities to participate in the global economy where segments of the populations lack verifiable identity modalities or have no access to banking services.

Network readiness
The Network Readiness Index measures the ability of the country to exploit the advantages offered by information communication and technology.Network readiness is closely tied to cryptocurrency adoption since it measures the preparedness of an economy in relation to the presence of infrastructure and skills required for technology adoption.It could provide guidelines to policy makers and ICT (Information and Communication Technology) stakeholders to collaborate and promote ICT development (Malisuwan et al., 2016).
The various factors to assess the impact on cryptocurrency adoption decisions discussed above are summarized in Table 2.

Model development and sample selection
Based on our above discussion based on what developmental factors impact CA, we present a model in this section with C as a dependent variable tied to the ten independent variables.These are democracy D, education E, Gini index G, GDP per capita GDP, human development HDI, corruption perception index CPI, regulatory quality RQ, control of corruption CC, economic freedom index EF and network readiness index N.Our model is as follows: Table 3 shows indices for democracy, education, income inequality, GDP per capita, human development, corruption perception, regulatory quality, control of corruption, economic freedom, and network readiness.The global crypto adoption index C, ranks countries on a scale of 0-1.The closer the score is to 1, the higher the rank (Chinalysis, 2020).The democracy index ranks 167 countries (Economic Intelligence Unit, 2019) scaled from 0 to 10.The education index and human development index rank 189 countries (Human Development Report, 2019) scaled from 0 to 1.The Gini index for income inequality ranks countries on a scale of 0 to 100 (World Bank, 2018) where a higher number means more inequality, the GDP per capita ranks countries according to their purchasing power parity (World Bank, 2019a).The corruption perception index CPI ranks 180 countries on a scale of 0 to 100 (Transparency International, 2019), where 0 specifies maximum perception of corruption and 100 specifies minimum perception of corruption.The regulatory quality index RQ and control of corruption CC scaled from 0 to 100 are a part of world wide governance indicators which rank over countries (World Bank, 2019b).High numbers express high regulatory quality index RQ and high control of corruption CC.The economic freedom index EF ranks 184 countries (Heritage Foundation, 2019) scaled from 0 to 100 and the network readiness index N ranks economies based on their performarnce across 60 variables (Portulans Institute, 2019).A high number means more economic freedom EF.
Table 3 shows the index values for 137 countries.For missing numbers, we apply the average numbers of the neighbouring countries as an approximation criterion.Afghanistan's Gini Index is assumed to be the average of Kazakhstan, Kyrgyzstan, and Tajikistan.Azerbaijan's Gini Index is estimated using its 2013 Gini index.The missing Gini Indices of Saudi Arabia, Kuwait, Bahrain, Qatar, and Oman are assumed to be the average of United Arab Emirates, Jordan, Iran, Qatar, and Yemen.Brunei's Gini and democracy index is assumed to be the same as UAE.Cambodia's Gini index is estimated using its 2013's Gini index.Hong Kong's control of corruption, regulatory quality, education, and Gini index is assumed to be the same as China's.Iraq economic freedom index is assumed to be the same as Iran's.Jamaica Gini index is estimated using its 2013 index.Maldives democracy index is based on its 2018 government restriction index.New Zealand's missing Gini Index is assumed to be the same as Australia's.Missing education index of Saudi Arabia is assumed to be the average of United Arab Emirates, Jordan, Iran, Qatar, and Yemen.Trinidad and Tobago's Gini index is assumed to be the same as Venezuela's.Uzbekistan's Gini index is estimated using the average of Kazakhstan, Kyrgyzstan, and Tajikistan.Venezuela's Gini Index is assumed to be the average of Brazil and Colombia.Yemen's economic freedom index assumed to be the average of Saudi Arabia and Oman.The network readiness index of Afghanistan is assumed to be the average of Kazakhstan, Kyrgyzstan, and Tajikistan.Network readiness index of Angola and Democratic Republic of Congo is assumed to be the average of Namibia and Zambia.Network readiness index of Benin, Togo and Burkina Faso is assumed to be same as that of Nigeria.The network readiness index of Bolivia is assumed to be the average of Argentina, Brazil, and Chile.Brunei's network readiness index is assumed to be the same as UAE's.Cape Verde's network readiness index is assumed to be the same as Senegal's.Chad's network readiness index is assumed to be the average of Nigeria and Cameroon.Gabon's network readiness index is assumed to be the average of Cameroon and Congo.Haiti's network readiness index is assumed to be the average of Jamaica and Dominican Republic.Iraq's network readiness index is assumed to be the same as Iran's.Network readiness index of Maldives is assumed to be the average of Sri Lanka and India.Montenegro's network readiness index is assumed to be the average of Croatia, Bosnia and Herzegovina, Serbia, and Albania.Myanmar's network readiness index is assumed to be the average of India and Bangladesh.Nicaragua's network readiness index is assumed to be the average of Costa Rica and Honduras.Papua New Guinea's network readiness index is assumed to be the average of Australia and Indonesia.Sudan's network readiness index is assumed to be the average of Chad and Egypt.Sweden's network readiness index is assumed to be the average of Norway and Finland.Uzbekistan's network readiness index is assumed to be the average of Kazakhstan, Kyrgyzstan, and Tajikistan.   A. Bhimani et al. which shows the importance of having prerequisite knowledge to operate the technology before its adoption.Network readiness also has a high correlation which again points towards the preparedness of an economy to the adoption of technology.The high correlation of CA C with human development HDI shows the willingness of economies investing highly in human welfare to continue facilitating their residents by upgrading their technology.It also shows an economy's willingness to adopt cryptocurrency to encourage financial inclusion and accelerate efforts towards identity management.A moderate correlation with the Gini index suggests that cryptocurrency could help lessen the income inequality by giving more opportunities to everyone alike.Weaker correlation exists with democracy, GDP per capita and regulatory quality.The corruption perception index CPI, control of corruption CC and economic freedom index EF exhibit a negative correlation which shows that countries with poor governance and economic structures would hinder the adoption of technology.Overall, the ten variables show a weak correlation with CA which could be attributed to the newness of the phenomenon and missing indicators which could affect the adoption.

Regression analysis
We regress cryptocurrency adoption C as the dependent variable and ten other independent variables with the regression line in Eq. ( 1).The hypothesis to be tested is whether a statistically significant linear relationship exists between the independent variables on the right hand side, i.e.D,E,G,GDP,HDI,CPI,RQ,CC,EF,N in Eq. ( 1), and the one dependent variable on the left hand side, i.e. cryptocurrency adoption C. The null hypothesis is that no statistically significant linear relationship exists, i.e. that the slope of the regression line is zero.
The first regression equation, tests for the relation of cryptocurrency adoption C as the dependent variable and democracy D as the independent variable.Table 4a exhibits the results of the regression The second regression equation, with cryptocurrency adoption C as the dependent variable, and democracy D and education E as the independent variables yields the results shown in Table 4b.
For the next regression equation, democracy D, education E and Gini Index G are chosen as the independent variables with cryptocurrency adoption C as the dependent variable.The results are demonstrated in Table 4c.
For the next regression equation, we add GDP per capita GDP as the independent variable in addition to democracy D, education E and Gini index G, where 0GDP means the number 0 multiplied with GDP (and analogously below).Cryptocurrency adoption C is the dependent variable.The results of the regression are shown in Eq. ( 5) and Table 4d.
For the regression equation Cryptocurrency adoption C is the dependent variable, and democracy D, education E, Gini index G, GDP per capita GDP and the human development index HDI are the independent variables.The results are shown in Table 4e.
For the regression equation Cryptocurrency adoption C is the dependent variable, and democracy D, education E, Gini index G, GDP per capita GDP, human development index HDI and corruption perception index CPI are the independent variables.The results are demonstrated in Table 4f.
For the regression equation we add one of the world governance indicators regulatory quality RQ as the independent variable along with democracy D, education E, Gini index G, GDP per capita GDP, human development index HDI and corruption perception index CPI.Cryptocurrency adoption C is the dependent variable.Table 4g shows the results.
For the regression equation we add another world governance indicator control of corruption CC along with other regulatory quality RQ, democracy D, education E, Gini index G, GDP per capita GDP, human development index HDI and corruption perception index CPI as the independent variables.Cryptocurrency adoption C is the dependent variable.Table 4h shows the results.
The regression equation  4j.

Structural equation modeling
To address the problem of multicollinearity among the variables, we use Eq (12) to run the structural equation model.The results are shown in Table 5a.
Next, we apply the variance inflation factor (VIF) test to address the problem of multicollinearity.The results are shown in Table 5b.
A variance inflation factor (VIF) of more than 10 means high covariance.We drop the variables with a VIF of 10 or more to produce the results shown in Table 5c.
To examine the difference between cryptocurrency adoption of developing and developed countries we introduce a dummy variable where 0 denotes developing and 1 denotes developed country.Economies in transition are categorized as developing countries for our analysis (United Nations, 2020).Applying the Generalized Method of Moments, we get the results shown in Table 5d.

Discussion, limitations and future research
Our analysis focuses on developmental indicators which are associated with receptivity to cryptocurrency adoption.The value of such an investigation is that this sheds light on what policy makers may focus on given the specific mix of developmental elements prevailing in their country context.We identify dimensions used by developmental economists encompassing the following: Education, the Human Development Index, the Network Readiness Index, the Gini index, Democracy, Regulatory Quality, and Gross Domestic Product, Control of Corruption, the Corruption Perception Index, and the Economic Freedom Index.These have been used in the prior literature in relation to aiding policy decision making capacity.While it must be borne in mind that a multitude of factors that are developmental in nature can impact technological operationalisation, we are able to focus on ones that are at the core of the discourse on development and national growth (Acemoglu, 2009).Decision makers and policy makers may appeal to data on these aggregate pointers which we suggest find positive or negative association.Little doubt exists that future studies will fine tune the data in investigating regions of the world, on different industry sectors and perhaps diversifying the methodology.Our study represents a first attempt to link cryptocurrency adoption vis-a-vis developmental parameters widely seen as tied to technology advances and economic growth at a macro-level.We have also argued, based on the recent literature, that cryptocurrency adoption can mobilise wider blockchain development and deployment in different country contexts.We report that, on regressing all the 10 variables together, GDP per capita GDP and corruption perception index CPI retains zero coefficient which shows that it does not affect the decision to adopt cryptocurrency in any way.The negative coefficient of the economic freedom index EF, democracy D and control of corruption CC and human development index HDI indicates that countries with liberal economic and regulatory laws, transparent governance systems and high human development may already have robust infrastructure in place and the cost of replacing those will be much greater than the potential benefits offered by cryptocurrency.The statistically significant relation with the Gini index G imply that greater income inequality is a motivator in adopting cryptocurrency to address the problem.The statistically significant relation with education E and network readiness index N is consistent with our expectation for them being the prerequisites for cryptocurrency adoption.
The p-values for control of corruption CC, regulatory quality RQ and  corruption perception index CPI are too high as shown in Table 4j.This may be due to strong correlation among the three variables as these three are indicators of governance quality.We repeated the regression adding variables in a different sequence.The results shown in Appendix 1 support the results of our prior regressions where we get at most education E, Gini index G and network readiness index N as the only statistically significant variables in any regression.
In our third set of regression analysis demonstrated in Appendix 2 we started by regressing all 10 independent variables and kept removing the variable with the highest p-value in subsequent regressions.This provided five statistically significant variables at most, i.e., education E, control of corruption CC, Gini index G, economic freedom index EF and the network readiness index N. Next, we undertook a country wise analysis excluding multicollinearity and addressing issue of endogeneity.We report a statistically significant relation with education E, GDP per capita GDP, economic freedom EF and network readiness N.However, the statistical insignificance of our dummy variable suggests that CA may not be country sensitive.Future research may conduct multigroup analysis after removing multicollinearity.Endogeneity issues that surface in studies like ours also may be addressed by researchers.
The statistical insignificance of most variables in our results could be attributed to the newness of the phenomenon and the limited understanding of its impact.Our cost benefit analysis reveals that newly industrialized economies from a wider pool of emerging economies, especially the ones with greater income inequality (Gini index) and possessing prerequisite infrastructure shown by the statistical significance of education and network readiness, will adopt cryptocurrency at a growing pace while economies with robust legal systems may be averse to the idea of wider adoption.
One limitation is the supplementation of missing numbers by using the average numbers of the neighbouring countries as an approximation criterion.That approximation is often viable but must be qualified.For example, assuming that Afghanistan's network readiness index N is the average of Kazakhstan, Kyrgyzstan and Tajikistan may be problematic given that Kazakhstan's index N is moderately high at 50.68.However, the estimated number for Afghanistan, 41.77, may not be unreasonable (see Serbian, 2021 on cryptocurrency mining in Afghanistan).Future research may apply other clustering methods as approximation criteria.
One potential limitation is whether the independent variables used in our study are necessarily independent, and whether the dependent variable can be independent.For example, a higher Human Development Index HDI may cause decreasing income inequality which could be a consequence of CA.More generally, causes and effects could be reversed.It is also possible that the independent variables may exist prior to or simultaneously with CA, which may support the notion that they are independent.Future research may address issues around independent/dependent variable linkages and other factors which might affect CA decisions.Future research may also consider analyses such as path analysis to statistically assess the direction of relationships, especially those that are statistically significant.

Conclusion
We developed a model to analyze factors in developed and developing economies which may accelerate or hinder the adoption of cryptocurrency and point to blockchain technology deployment as a consequence.The ten factors used in our study are drawn from prior studies addressing issues of technology adoption and their impact on nations at different degrees of development.Our results suggest that cryptocurrency usage can vary across nations given country specific factors tied to the level of economic development.Country based adoption decisions are guided by various factors outside simple technical issues or commercial value impact assessments.While current research has mainly focused on technical aspects of cryptocurrency deployment, a need exists to further research wider level developmental factors which our study results point to.
The study results are of value to both policy makers and researchers.Our results confirm, like other broader studies on Information and Communication Technology, that differentials across developing and developed countries impact CA (cryptocurrency adoption) in relation to macro-national developmental factors.We report a positive relationship with educational level, the Human Development Index, the extent of network readiness, the Gini index magnitude, the level of democracy, the extent of regulatory effectiveness and the Gross Domestic Product.These are clear indicators that leapfrogging is a possibility if national decision makers focus on initiatives and investments that enhance these factors.Further, the results provide a basis for prioritising economic and social programs of focus for governments.We report also that the Corruption Perception Index and the Economic Freedom Index are inversely associated with CA.This further enhances the policy prioritisation perspective offered to policy makers.Policy makers in emerging economies may benefit from the results presented here by assessing those associations in the 137 countries which bear relevance to their nation or region where similarities to developmental indices we have explored are in place.The factors outlined may have a cause effect relationship which are country specific, and which create particularity in terms of cryptocurrency adoption.Such a focus directs decision makers to elements of governance, human development and inequality indicators that may be prioritized.As in developed countries, CA may be a precursor to wider blockchain technology-based platforms becoming operationalised.

Regression Results
Results from Structural Equation Modelling.

Declaration of Competing Interest
The authors declare no competing interests.

Funding
No funding was received.
Table 1 demonstrates the Pearson Correlation between cryptocurrency adoption C, Democracy D, Education E, Gini Index G, Gross Domestic Product GDP, Human Development Index HDI, Corruption Perception Index CPI, Regulatory Quality RQ, Control of Corruption CC, Economic Freedom Index EF and Network Readiness Index N. Cryptocurrency adoption C has the strongest correlation with education E

Table 2
Summary of variables and their effects.

Table 3
Variables data for 137 countries.
(continued on next page)

Table 5a
Table 4e-j shows that the human development index HDI also gives a very high p-value.The pvalue for democracy D is quite high, i.e., 0.55 in the simple linear regression with one independent variable.It declines to its lowest of 0.147 when education E and the Gini index G are introduced as shown in Table 4c and increases to as high as 0.743 in Table 4g.The p-values for most independent variables are quite high.The highest p-value of education E is 0.151 but it drops to a low of 0.001 in Table 4d.The Gini index G and network readiness index N show the lowest p-values of all the variables.