Babysitting Bitcoin

Getting your Trinity Audio player ready...

Blockchain explainers often begin with Alice & Bob communicating transactions through a ledger. In contrast, economics traditionally starts with the parable of Robinson Crusoe, isolated from the rest of humanity, armed merely with his time, skills, and a few scarce resources to serve his desired ends.

The Crusoe parable may be useful in introducing the concepts of scarcity, constrained optimisation, and time preference, but it misses the fundamental problem at the heart of economics as applied to society as a whole: the coordination and exchange of effort using money-things as an intermediary.

Babysitting as a parable

A better parable to connect real-world economics and blockchain is the Capitol Hill babysitting co-op. The co-op was a real thing, started in Washington D.C. back in the 1950s as a way for Capitol Hill staffers to coordinate reciprocal baby-sitting service through a central intermediary.

The story was originally told in the Journal of Money, Credit & Banking in 1977 by two members of the co-op, Joan and Richard Sweeney (the latter also happened to be deputy director in the Office of International Monetary Research at the U.S. Treasury). It has been retold numerous times over the ensuing years, most famously by the Nobel Prize-winning trade theorist, Paul Krugman suggesting “it will change the way you think about the world” (some may say arguing outside his lane, but, hey, according to him it “changed my life”).

HODLers, not toddlers

In short, the story is that paper scrip (tokens) were issued to be used as money for the families to spend on baby-sitting. It worked for a while, but people wanted to hold more of these token promises, just in case they’d spent their earnings but needed a baby-sitter urgently: essentially a precautionary demand for money to cover for uncertainty. The central administrative authority charged a fee (a tax effectively) to pay administrators (government spending), but this was too big in relation to the tokens available because people also wanted to hoard tokens. One member has been quoted as saying that the tokens were in such demand that “you would kill for scrip … you would sell your children for scrip”! HODLers for sure! The imbalance set in motion a cycle of action and reaction which destabilised the system.

A baby-sitting recession ensued as a result of what may be recognised as the Paradox of Thrift (an emergent property of the economic system, consistent with rational human action but collectively irrational): no one wanted to spend their precious tokens. To end the recession, the central administrators simply ordered people to go out more and baby-sit more! Dirigisme failed, so they just printed more tokens and distributed them to households (a deficit-financed giveaway).

There was a short boom, but an inflationary recession developed as members had too many tokens hence no-one needed to work (baby-sit) anymore. The tokens were like hot potatoes, but they couldn’t spend them. Presumably some members simply circumvented the ‘government’ system and reverted to a form of bilateral credit-based barter/reciprocation, but a monetary bust-boom-bust cycle was created even though there was no fundamental ‘real’ shock to the economy of the co-op. Far from being the mythical veil specified in orthodox economic theory, money itself was the problem.

Lessons for macroeconomics

The co-op is a great intuition pump to get your head around complex problems in monetary macroeconomics. Various remedies have been suggested: Keynesian oriented economists such as Krugman suggested the central administrator could lend tokens at interest, or the community could start a secondary market in tokens (a kind of peer-to-peer capital market), or even add a form of demurrage—negative interest rates—to stop people hoarding; Austrians suggest the price of babysitting should simply be made more flexible (for instance, allow ‘½-hour of babysitting’ to be a kind of variable unit of account where it may in fact only buy 15 minutes of babysitting), or adjust for peak-time usage (there actually was a form of peak-time pricing).

All of these approaches have merit, nicely demonstrating the perennial tension of the inflate-versus-deflate perspective generally held by Keynesian and Austrian economists. However, there is a fundamental issue that has generally been overlooked: a book-keeping credit system is qualitatively different to a token money system.

Inside money & outside money

In the early days, the co-op operated on a bookkeeping system (aka mutual-credit clearing). Originally, with relatively few members, this mutual-credit clearing system was easy to administrate, and the implicit quid pro quo nature of exchange was understood: you don’t need to hoard babysitting credits because you always (within reason) have access to short-term credit. However, as the system grew, the ledger became too onerous to maintain.

As the original paper by Sweeney & Sweeney argued, tokens may be useful as they reduce the work and potential for errors in the ledger. The administrators just handed out Monopoly™ money, but while this administrative expediency appeared a simple solution, it represented a fundamental change in the nature of the economy: in economist jargon, flexible endogenous credit money (book-keeping, ‘inside money’) was replaced by an arbitrarily fixed exogenous money (tokenized, ‘outside money’). 

When the tokenized exogenous money was issued, it broke the link between the supply of effort and the supply of money, and appeared to disembody the mutualist nature of the community. The token issuer had far more agency in determining the system, not because the token issue was fiat in nature (the ‘fiat’ unit of account was simply ½-hour of baby-sitting time), a commodity would have presented its own problems. No credit was available: there was no flexibility in the system.

Lessons for distributed ledgers

The babysitting co-op has been much debated and various lessons for various disciplines can be taken from this parable, too extensive to go into great detail here. One pertinent lesson is that a distributed ledger can precisely serve the purpose of community mutual-credit clearing of the kind that proved too problematic to administer back in the 1970s. The transactions could have been validated and logged by mutual consent with no additional work; the books would always balance. Dishonesty would always be discovered and ultimately rooted out.

Alongside a real-time public ledger of activity, other aspects of the system design could likely be automated: algorithmic contracts with automated credit limits and penalties for hoarders; centralisation of booking; a centralised market for bids and offers for baby-sitting; real-time settlement etc. It’s actually a simple example of how technology can make life a lot easier in a market-based economy and reduce the requirement for onerous centralised management of some aspects of economic activity: it provides valuable information (beyond mere prices) to help the market coordinate activity.

Money coordinates scarce resources and abundance

Scarcity was the principal problem faced by Robinson Crusoe (and humanity for thousands of years). Resource constraints remain a major concern for different reasons today, yet much utility is served through information rather than physical resources. The key issue is to coordinate the distribution of abundance while maintaining incentives to produce the abundance. For good or ill, money—both as a unit of account and quantifiable money-things—will remain the primary coordinating mechanism. This can create problems for no other reason than a monetary system can be badly designed, and/or badly implemented.

Back in the 1970s, while Friedrich Hayek was writing about the Denationalisation of Money, the post-Keynesian economist Hyman Minsky pointed out that while anyone can create ‘money’, getting it accepted is the difficult part (ask BTC promoters working in El Salvador). Blockchain technology now makes the process of creating money-things simpler and more efficient: literally anyone can create ‘money’! Various tokens can be created to represent either some fungible money-thing or otherwise exchangeable value within a particular community. They can be used to both satiate need and coordinate activity within those communities. But other than ‘persuading’ dictators to impose them on their people, how do these money-things become accepted?

A narrow form of money-thing only has to serve within its own community: it does not need to be a universal money in order to be a form of money. But the more such money-things expand beyond a special purpose (such as within a game or a babysitting co-op, or even a cult), the more demand/acceptance is created, the greater impact of a network effect, and ultimately the more moneyness these tokens contain. They begin to look like ‘money’ (let’s ignore the unit of account debate for now).

Different monies, different purposes

Different communities may require different types of money-things. Where deflation is both possible and desirable in a particular community, for promoting competition, such as for electronic transaction processing, it may well serve that community best that the money-things are limited in supply, and more likely to maintain value or appreciate (as long as the community sees value within that particular economic system). Inefficient suppliers will eventually be priced out of the system by competitive elimination.

Other communities may have less reason to desire such competition and deflation. A people-based cooperative, such as the babysitting co-op, or a community time-bank are cases in point: having a money that endogenously adjusts in quantity based on the demand and supply of the underlying economy makes more sense where nominal growth is the norm and the restoration of equilibrium through timely adjustable pricing or monetary velocity becomes problematic. Flexible credit money is more desirable than a form of commodity money with fixed or limited supply. Regulatory systems to maintain the resilience of the credit system (e.g., not allowing long-term hoarding of credits, or excessive creation of units) are crucial to maintaining stability.

A universal money, or a multi-currency world?

The most complicated type of money to design is a universal currency. Whether one likes government policy actions (errors?) to be inflationary or deflationary is a political choice, not an economic one: inflation and deflation benefit or penalise various parts of society differently. This is why history has meandered through various attempts at commodity money, credit money, and government fiat money: often all together in one large system, often failing in their own particular way, to be re-engineered for a particular context. It’s a ‘wicked problem’ which won’t be solved outside politics by economists alone, let alone ‘Bitcoin’ as a currency as the BTC maximalists seem to think.

The Bitcoin blockchain in its role as an extraordinarily efficient information-transaction ledger and tool for tokenization of various things of value to create types of money-things can certainly be crucial in helping coordinate and engineer a global, multi-currency monetary economy that is more suited to the needs of the 21st century. From private currencies and stablecoins to CBDCs, carbon-credits, community time currencies, NFTs, gaming-tokens and beyond, the global monetary system is now truly open-season for monetary designers.

Inasmuch as there is utility, the underlying native currency of this transaction system will also generate its own value within the context of its community: an information commodity for market-based coordination. But dreams of a ‘universal money’ should be seen for what they are. Forget Robinson Crusoe, the babysitting co-op is a much more useful tool for thinking about economics.

Watch: CoinGeek Zurich panel, The History of Money & The Future of Bitcoin

https://www.youtube.com/watch?v=_clPaClcZoE