Price Dynamics Of A (Emerging) Digital Bridge Asset

Cryptocurrencies are experiencing unprecedented momentum. While Bitcoin & Ethereum have been extensively (and controversily) discussed, the best-performing (and third largest digital asset by network value) this year, XRP, hasn’t been much in the spotlight. The usecase (‘digital bridge asset’) seemed unique and caught my attention. I wanted to understand if and how this usecase could prove to be valueable for the asset- and share my findings with you. You’ll find some basic background information/ introduction on XRP at the very end of the article.

So how will the process of a bridged payment look like?

Let’s imagine a foreign worker who instructs a remittance (JPY 30,000) from Japan to his family in Thailand (in THB). Bank A in Japan to route the payment instruction through Ripple‘s network (RippleNet) to Bank B in Thailand. A currency conversion needs to take place, so market makers (quoting exchange rates) are needed.

Let’s use a simplified model, in which market makers are quoting the following markets:

Pathfinding within RippleNet would identify two options:

A) Sell JPY 30,000 for USD 271.59 & sell USD 271.59 for THB 9,038.57 
B) Sell JPY 30,000 for XRP 1696.84 & sell XRP 1696.84 for THB 9,039.59

We can see that the bank would save THB 1.02 for this payment by bridging via XRP. Assuming FX costs to be THB 3.13 (= the difference between actual USD bridging and theoretic USD mid quotes, ie 0% bid/offer spread), XRP reduced Bank A’s FX costs by ~32.7%. This cost saving is, partly, being ‘sponsored’ by Ripple. Ripple is subsidizing market makers in the beginning, until sufficient liquidity and network effects in a given FX corridor have been built. The increased usage (and value) of XRP should compensate Ripple in the long run, when subsidies won’t be neccessary anymore (similar to all other currencies/assets that market makers are quoting).

So even within Ripple’s highly efficient network, XRP can be favourable vs USD. Not to mention that XRP enables the payment to settle instantly while USDs would still need to be transferred the traditional way, via SWIFT. I’m skipping a cost comparison vs SWIFT, as it is charging banks a magnitude of fees/ FX costs, with high error rates and days to settle.

Now the question is, how does this successful bridging affect XRP’s price?

To answer this question, let’s have a quick recap on the basics of market making: A market maker quotes a market (ie a bid, and an offer). If he gets ‘hit’ on his bid, he knows that he has likely been the highest bid in the market. In order to adjust to the market (and hopefully ‘overpaying’ less on the next trade), his new market (ie both bid and offer) will be adjusted lower, slightly. This makes perfect sense, because when people are selling (‘hitting bids’), prices go down. Same pattern in case people buy (‘lifting the offers’), the market maker will adjust his market higher, as he’s likely been the cheapest offer in the market.

Below you will see the orderbooks (showing all bids and offers from the market makers) for every given market, and how they’ll adjust in our payment simulation:

The XRP/JPY market maker lost some of his XRP, and now holding more JPY. Quoted size on the bid side increased, as he now has some more JPY to sell in order to get back to his original XRP holdings. His market remains unchanged (some market makers might already adjust their market), as the payment size < quoted sizes. The XRP/THB market maker holding a bigger XRP position now, while loosing some of his THB. We can see that for the XRP/JPY markets, the orderbook has become a bit more ‘XRP bullish’ (bigger sizes on the bid side), whereas XRP/THB orderbooks became a little more ‘XRP bearish’. XRP/USD markets remain untouched. 
Overall, there’s no price effect to observe in all markets.

Now let’s imagine another payment coming in, JPY 265,000 (~$2,400 or XRP 15,000) from a Japanese corporate paying its Thai supplier:

Our XRP/JPY market maker’s offer got lifted, even the market maker with the second best offer (given the bigger size of the payment). Bid sides have been adjusted higher (by around 0.5x bid/offer spread, to meet the prior bid/offer spread). The opposite is true for the XRP/THB pair.

We can now observe a price impact (with XRP vs JPY trading up, and XRP vs THB trading down).

What happened? Both market makers (partly) missed out on this trade. The payment volume exceeded their quoted size. No trade -> no bid/offer spread to earn. In our simulation, the XRP/JPY market maker has just experienced net XRP buying from Bank A (a market taker). Despite adjusting the bid side higher, the market maker isn’t able to get back to his original XRP position. He needs to become a market taker, so that he has a chance in catching the next payment(s) in this FX corridor. In our simplified example, there are 2 ways for him to do so:

A) Buy outright XRP vs some of the USD he’s holding. To regain his XRP 16,697 he’d need to pay a (weighted average) $2,672.13 
B) Buy outright XRP vs THB, as XRP supply in this market should have made XRP ‘cheap’. He’d need to convert his USD into THB (@ 33.28) and spend (an equivalent of) $2,672.78 to get to his XRP 16,697

For now, the market maker is better off with option A. The ‘cheap’ XRP from the XRP/THB markets will need to soften a bit more to make the arbitrage work, and take out the XRP/THB market maker. After executing on option A, the XRP/USD markets are now up 0.02%, with a stronger bid side building up:

This is the interesting part: we just saw that the XRP/JPY market maker actually reduced some of his ‘traditional’ bridge asset holdings (USD), in favour of adding XRP. As payments are taking the path of lowest cost, market makers need to make sure to not run out of XRP.

After having examined the impact of 2 smaller payments, in a simplified model, it was demonstrated that higher payment volumes result in higher demand for the cheapest bridge asset. It should also make sense for banks to directly hold a stake of XRP, in order to ‘save’ 1x bid/offer spreads (in our example: just pay the bid/offer spread for the XRP->THB exchange, and not the JPY->XRP).

XRP dynamics in the ‘realworld’ are expected to be significantly more complex, but likely also more impactful:

— XRP is constantly improved, can already handle ~1,500 transactions every second (and up to ~50,000 per second off-ledger) 
— there might be hundreds of market makers & banks quoting bid/offers, in up to 180 currencies that the UN recognizes as legal tender 
— it is not clear (yet) what instant settlement and drastic cost reductions will do to payment volumes (currently $155tn annual B2B cross border payments)
 — speculation: early investors might take profits, whereas market makers/banks/speculators might engage in XRP outright buying & holding, to frontrun/take advantage of what we just demonstrated
 — the most liquid G7 corridors (like EUR/USD) might be (too) expensive for Ripple to tackle, it might make more sense to focus on Emerging Market currencies to build the initial network effects

Thoughts & comments welcome.

Background information: XRP’s mission is to become a universal, digital bridge asset. A quite unique usecase, and a neccessary step to enable the internet of value (where money moves the way information does today). The maximum number of XRPs in existence is 100bn units (as per protocol, similar to Bitcoin), and all of it has been created in 2013 and gifted to a company called Ripple. Ripple still holds ~60% of all XRP (valued at $10bn+) and uses its XRP stake to incentivize usage and grow the XRP ecosystem. The company is backed by banks, Google, prestigious VCs and has been growing to 150+ employees with more than 90 high profile banks/customers and 75+ commercial implementations this year. A more detailed writeup on Ripple & XRP can be found here.