Money as Bookkeeping
The three common ways of theorizing money — as a commodity, legal construction, and credit — can be effectively reconciled by considering it as a social institution for bookkeeping. This article outlines the four perspectives, lists materials for understanding each, and considers how some recent innovations in digital record-keeping fit into the picture.
According to this view, the nature and value of money is best understood by considering how certain relatively scarce commodities have historically emerged as general mediums of exchange, with precious metals and especially gold seen as paradigmatic. A particular commodity can acquire monetary functions if it is durable, fungible, difficult to produce and counterfeit, easily divisible, and widely accepted as a good store of value.
The fact that it is possible to have a functional monetary economy without a commodity standard makes this view problematic as a general explanation of money. However, it remains implicit in theories that view the economy as fundamentally a system of bartering in which money is merely one commodity among others. It also serves as the primary source of ideological inspiration for those who find attractive the idea of “sound money” and wish to constrain the monetary powers of public authorities and financial institutions.
According to this view, money is a “creation of law,” established by public authorities as the official money of account and that which can be delivered and must be accepted to settle money-denominated contracts. Taxes and other state-defined obligations are seen as the initial drivers of demand for government issued money, which circulates in the economy as uncollected tax credits, equivalent to the net financial savings of the non-government sector. This view has recently been popularized by neo-chartalists as part of what has become known as Modern Monetary Theory (MMT).
Of course, in addition to money issued by governments, contemporary economic systems also include large quantities of money issued by private financial institutions, which brings us to the third common perspective.
According to this view, money is a system of credit with various monetary instruments serving as representations of credit relations. This is most evident in the case of banks issuing loans which create deposits that are then used as a means of payment . But this approach can also be generalized to include government issued money, understood as credit redeemable through taxation. In the context of the credit view, it is thus crucial for the value and survival of both state and bank money that their issuers accept it back at nominal value. 
In so-called fiat monetary systems with floating exchange rates and modern banking practices, pro-cyclical credit creation introduces the possibility of nominal expansion outpacing the capacity for real economic growth. Ideally, the financial system should direct credit money toward productive economic purposes, thereby enabling material progress and innovation. In reality, credit is often not used for new capital formation, but ends up inflating the prices of existing assets instead.
For some, trying to prove which of the three perspectives is “correct” is a matter of great importance; for others, the relevance of each view depends on the theoretical and historical context; and still others ignore the topic altogether while operating under a variety of implicit assumptions about what money is and how it relates to the economy at large. In any case, each of the three perspectives focuses on a different aspect of the same phenomenon, and anyone interested in learning more about how they relate to each other should read the materials referenced above.
What I would like to highlight for the purposes of this article, however, is that all three perspectives can be reconciled within a fourth:
4. Bookkeeping view 
According to this view, the monetary system is a collective method of enabling and keeping track of economic activity and relations using a socially agreed upon measuring unit. In other words, a bookkeeping system that allows individuals, organizations and society at large to value assets and activities, mobilize real resources, clear and settle transactions , carry into the present their financial history, and project into the future their economic plans and interdependencies. As such, it represents a form of communal, institutionalized memory , regardless of whether its rules are officially mandated or based on a set of informal norms and practices.
The exact nature of and limitations on the production and use of monetary instruments are contingent on legal, cultural, and other historical circumstances. A monetary community could theoretically agree upon any number of arbitrary rules on when, how, and by whom various accounts can be credited or debited, what can serve as an acceptable means of payment , and how the production of currency is regulated. In its most general formulation, the bookkeeping view is agnostic on such details, while admitting that different systems vary in their real-economic effects and implications.
In terms of economic analysis, the bookkeeping view emphasizes the need to accurately discern and interpret the asset/liability structure and financial stocks and flows recorded on balance sheets and ledgers throughout the economy, and their relationship to real-economic production and distribution. This approach can be combined with agent-based methodologies for an even more complete view of economic processes. 
Digital, Decentralized and Cryptographic Bookkeeping
Depending on their rules of issuance and governance, specific digital and cryptocurrencies may more easily fit into the commodity, credit, or even state view. But all are also systems of bookkeeping for the unit(s) specified by their respective protocols or issuing authority.
Defining and issuing bookkeeping units in the context of some economic activity or contract is nothing special. Indeed, anyone can do it at will. But good luck getting others to accept your newly defined units as attractive stores of value or means of payment. Such privilege has historically been preserved for large and powerful institutions, even though access to the monetary system has been democratizing considerably over time.
The reason why completely arbitrary units in some cryptographically secured digital bookkeeping systems have acquired a little bit of private “moneyness” has a lot to do with certain novel features and capabilities of their underlying protocols and networks. These include decentralization, relatively high privacy guarantees, censorship-resistance, improved auditability, open and global access, and programmability. These qualities have attracted loyal communities and driven economic activity, with bitcoin and ether in particular experiencing commitment from users to think of these as money.
Importantly, the emergence of these systems has triggered new ideas and activity in areas such as central bank  and corporate digital currencies , as well as a slew of initiatives aspiring to restructure the currently highly concentrated business of managing not just financial but all kinds of information . Combined, these efforts represent a major — although still a nascent — trend in the evolution of the administrative structure and logic  of the global digital economy.
Economics inspired by these new techniques of information management will remain marginal as long as crypto-financial networks and services struggle to achieve mainstream adoption. The need for analyses based on the bookkeeping view of money, and the ability of crypto-finance to attract the attention of people capable of providing it, will depend on two key drivers:
(1) How much real capital development (both in terms of the deployment of distributed network infrastructure as well as software and services running on top of it) and exchange is facilitated through various crypto-financial instruments. In other words, the size of the real economy enabled and tracked by these new forms of record-keeping;
(2) Degree of integration between traditional and crypto-financial systems.
In a mass adoption and integration scenario, many of the well-known problems of monetary and macroeconomics are certain to reemerge, although possibly in a context of unprecedented global scale and considerable institutional reconfiguration between nation states, corporations, and decentralized information networks with varying degrees of geographic reach and sovereignty . But if that happens, the bookkeeping view of money will be an essential tool for (crypto-)economic analysis, regardless of whether the monetary instruments in question are digital commodities, defined and regulated by (de)centralized authorities, or issued as credit.
 Menger, C. 1892. On the Origins of Money. Economic Journal, Volume 2, Issue 6, pp. 239–255. Available here. (Originally published in German in 1892.)
 Jevons, W. S. 1896. Money and the Mechanism of Exchange. D. Appleton and Company. Available here. (Originally published in 1875.)
 Mises. L. 1953. The Theory of Money and Credit (especially Parts I and II). Yale University Press. Available here. (Originally published in German in 1912.)
 Knapp, G. F. 1924. The State Theory of Money. Macmillan & Company. Available here. (Originally published in German in 1905.)
 Wray, L. R. 2012. Introduction to an Alternative History of Money. Levy Economics Institute Working Paper №717. Available here.
 Wray, L. R. 2014. From the State Theory of Money to Modern Money Theory: An Alternative to Economic Orthodoxy. Levy Economics Institute Working Paper №792. Available here.
 Innes, A. M. 1914. The Credit Theory of Money. The Banking Law Journal, Volume 31, Dec./Jan., pp. 151–168. Available here.
 Schumpeter, J. A. 1949. The Theory of Economic Development. An Inquiry into Profits, Capital, Credit, Interest, and the Business Cycle (in particular Chapter III). Harvard University Press. Available here. (Originally published in German in 1911.)
 Ingham, G. 2004. The Nature of Money. Polity Press. Summary article available here.
 This is explained by the endogenous money theory by which bank deposits are created through the issuance of loans (“out of thin air,” as the popular saying goes), regardless of whether the issuer has sufficient central bank reserves at the time of the loan as these can usually be obtained ex-post. It also exemplifies the usefulness of discussing money in conjunction with balance sheets. See for example McLeay, M., Radia, A. & Thomas, R. 2014. Money in the modern economy: an introduction. Bank of England Quarterly Bulletin 2014 Q1. Available here.
 Tymoigne, E. 2017. On the Centrality of Redemption: Linking the State and Credit Theories of Money through a Financial Approach to Money. Levy Economics Institute Working Paper №890. Available here.
 I initially called this fourth perspective “the accounting view” but Jan Kregel suggested that it is better known as the “balance sheet” or “bookkeepers” view which is historically prior to Knapp, sees money as simultaneously a credit and a liability, and puts an added emphasis on the clearinghouse concept. This makes it naturally applicable to distributed ledgers and payment networks in which clearing and assets are generally fused (as in Bitcoin) although should be separated analytically. While commodity money is often described as “nobody’s liability”, it still represents “credits” in relation to anyone selling goods or services for money. As Alfred Mitchell-Innes (1914, referenced above) put it: “By sale a credit is acquired, by purchase a debt is created.” Commodity money can thus be seen as fundamentally compatible with the bookkeeping view in the sense that a monetary system based on a commodity standard is still a social institution for keeping count of the distribution of money among its users, even though the accounting balances are not necessarily visible or represented in a centralized or distributed ledger.
 This is done through payment systems which are subject to very strong network effects and — unless closely regulated — offer system operators various lucrative privileges. Indeed, questions regarding control over the dominant payment system in a digital economy are at the center of contemporary monetary thought and regulation.
 I am indebted to Alex Evans for suggesting “communal memory” as a way to describe this one important function of the monetary system as bookkeeping.
 The actual means of payment, the “money things” such as cash, are popularly perceived as the money, or “money proper” as J. M. Keynes called it. This can and has taken a variety of forms, including metallic coins and little pieces of paper. More recently, bank money expressed in the official money of account has emerged as a widely used substitute for state-issued cash, primarily because much of it is guaranteed redeemable in the latter. That said, transaction clearing by centralized bookkeepers without the use of physical money is not unique to contemporary capitalism.
 Bezemer, D. J. Towards an ‘accounting view’ on money, banking and the macroeconomy: history, empirics, theory. Cambridge Journal of Economics, Volume 40, Issue 5, pp. 1275–1295. Available here.
 Barontini, C. & Holden, H. 2019. Proceeding with caution — a survey on central bank digital currency. Bank of International Settlement Paper №101. Available here.
 See Placeholder’s portfolio for examples.
 I have previously described the emergence of decentralized blockchain networks as an important step in the evolution of bureaucratic organization — an institutional innovation with potential to accelerate and amplify administrative globalization and automation.
 See Joel Monegro’s article on Sovereign Cryptonetworks.