Digital Gold / Digital Cash Network Effects

This post is in response to Kyle Samani’s “On the Network Effects of Stores of Value”. Kyle’s post begins with an explanation of network effect concepts, followed by applications of those concepts to common crypto-money narratives. I would encourage you to read his piece before reading my own.

Kyle’s thesis is that “digital cash” has a better network effect than “digital gold”, and thus an asset vying to become “digital cash” has a higher expected value than one vying to become “digital gold”. Kyle assumes that Bitcoin is vying to become “digital gold”. What we end up discovering is that Kyle’s definition of the “digital gold” narrative is just an incomplete definition of money, and his definition of the “digital cash” narrative is just a payments network.

Digital Gold

The metaphor of Bitcoin as “digital gold” is deeply ingrained, though there is no singular definition of what that metaphor means. There are at least two views on its meaning. One is that “digital gold” will be like what gold is today in Western economies: an asset which is almost never used to make payments and is marketed to speculators as a way to hedge against adverse macroeconomic scenarios. A second definition is that “digital gold” will be like what gold was in the 19th century: an international standard of value which is regularly used to make payments — a global money.

Kyle introduces a third definition for the “digital gold” narrative…

The purpose of a store of value such as digital gold is to… store value for consumption at some later date. Other than the time in which the digital gold is converted into something else, digital gold just sits there, doing nothing.

Why would an individual or business use “digital gold”, a non-productive speculative asset, to store value for settlement of future liabilities? Individuals and businesses are generally looking to match the duration and denomination of their assets to their liabilities. The future consumption is a liability not denominated in the “digital gold”, otherwise we’re stating that the “digital gold” is a medium of exchange, thus piercing the “digital gold” narrative and entering the “digital cash” narrative. The reason this definition of “digital gold” as a “store of value” seems incomplete is that it either omits extenuating circumstances (the “digital gold as a macro hedge” definition) or it is trying to look at one monetary property, store of value, in a vacuum.

When a user wants to liquidate their digital gold to consume some other good or service, she needs to find liquidity: someone who is willing to purchase the digital gold. This can be done at an exchange that specializes in fungible digital gold.

Again, we’re left wondering why, in this narrative, did the user go through the Rube Goldberg machine of buying “digital gold”, sitting on it, and then selling it. It would have made more sense for the user hold their money or to lend it out for a return. Something is missing.

Digital Cash

What was missing from Kyle’s definition of “digital gold” can be found in his definition of “digital cash”…

The purpose of digital cash is to both store value and be used as a medium of exchange. Additionally, digital cash can become a unit of account.

Here we just end up with the basic definition of money.

The utility of digital cash in aggregate therefore is a function of how many merchants want to accept payment for goods and services in digital cash.

If there is an intermediary payment processor who receives the “digital cash” and deposits USD in the merchant’s bank account, aren’t we back to the “digital gold” narrative of having to find liquidity? For the “digital cash” narrative to stay differentiated, we have to assume that merchants want to put the “digital cash” on their balance sheet. In practice, they don’t. Even if some employees and suppliers want to be paid with “digital cash”, their contracts are USD-denominated, so you would still minimize or hedge its presence on the balance sheet.

This resembles the direct network effect (the telephone diagram) as described above. The more people who accept payment using digital cash, the more merchants existing users can do business with.

This assumes that a user can not do business with a merchant if they don’t accept the user’s “digital cash”. Sometimes this assumption is true, and the few merchants who do accept “digital cash” do so because it is incremental revenue, not because it is convenient or cheap. However, this assumption is generally not true, users have a diverse set of payment options that they can fallback on if their preferred one is not available. The payments industry is saturated. This means that the payments’ network effect is limited to niche use cases. “Digital cash” is a small S-curve.

Lastly, the “digital cash” network effect dilutes itself out, there is a very low marginal cost for merchant payment processors to add a multitude of “digital cash” options. The merchant is ultimately indifferent because they have USD deposited in their bank account, whether it was converted from Dogecoin or Dash.

The “digital cash” narrative collapses from a very promising definition of “money” to the old “merchant adoption” fallacy — there is no path from “accept for payment” to “holding as money”. There is, on the other hand, a path in the reverse. If a business owner wants to accumulate “digital gold” as a speculative bet then it may be less expensive for them to do so through payments to their business than by going on an exchange. Ultimately though, merchant adoption will happen due to Thiers’ law: good money drives out bad.

Kyle also describes Chartalism, but that is worth a separate post.

Digital Money and Value

if something else becomes digital cash with a super-linear network effect while Bitcoin remains digital gold with a sub-linear network effect, then Bitcoin will be surpassed

A super-linear network effect in a small total addressable market (ersatz payments) is not going to surpass a sub-linear network effect in a massive total addressable market (money).

The SoV view is based upon reflexivity: The more people hold it, the more valuable it becomes, therefore driving more people to hold it. […] But when the price decreases, the underlying value of the utility value creates an organic price floor.

I think the reflexivity is actually more pronounced when the price decreases: the underlying value of the utility value decreases. Liquidity is drying up and it becomes increasingly uneconomical to use it as a method of payment. What creates the organic price floor is long-term speculators who have the conviction to continue accumulating through the volatility.

The Liquidity Moat

Bitcoin will continue to dominate exchange and OTC liquidity until the credibility of its monetary policy is jeopardized.

From Bloomberg
Bitcoin has a lead with CME, CBOE, and NASDAQ futures. But given Ethereum’s trajectory, it seems quite reasonable that it will achieve parity within 24 months.

Parity is only reasonable if we assume that new Bitcoin financial instruments will not be introduced within 24 months.


The battle to be the super-mega winner of crypto is just beginning.

I’m skeptical that that’s the case, it seems to be over. There will be a constant churn among small assets with ethereal super-linear network effects, but Bitcoin is continuing on its multi-decade, behemoth sub-linear network effect as the premier global money.