Liquidity Monitoring: Depth

In our last liquidity monitoring post, we introduced the concept of dislocation as a way to measure the price competitiveness of an XRP-fiat pair. In this post, we introduce the companion depth metric and combine both metrics into a data visualization for assessing liquidity performance.

Depth

Dislocation tells us how competitive an exchange’s XRP prices are, but it ignores the important quantity component of liquidity. Recall that the executed price against an order book is size-dependent. Denser order books give more competitive pricing for larger orders than thin order books. We want to know how large a payment an exchange can support before prices get significantly worse.

Fortunately, this is a bit easier to define than dislocation–we simply look into the order book and count up how much XRP is within a fixed percentage away from the mid-price. The caveat here is which side of the order book to look at. For ODL, some XRP-fiat pairs are typically only destination (source) corridors, so we can focus on the bid (ask) side of the order book. For XRP-fiat pairs that are both source and destination, we calculate depth for both bid and ask sides of the order book, then take the more conservative result.

Depth is then a measure of how much XRP is available “close” to an order book's best price. It compliments dislocation by providing a measure of the liquidity capacity of an exchange. If dislocation is the zero-order description, this is a first order correction that considers the size of the payment.

Depth helps inform payment execution strategies for ODL, highlighting the trade-off between the immediacy of larger payment sizes and the quality of executed price. And like dislocation, any change in typical depth is worth investigating to understand its impact on ODL.

Visualizing

Importantly, depth and dislocation are measuring different things. Dislocation tells you how good or bad prices are and depth tells you how large a payment you can send close to those prices. Even more importantly, depth and dislocation are readily comparable across exchanges and XRP-fiat pairs.  

Inspired by Hans Rosling and Gap Minder, we can visualize this data in a dislocation versus depth plot. As seen below, the x-axis shows median daily depth within 100 basis points of the quote at an exchange. The y-axis shows median dislocation in basis-points from the reference index price for that XRP-fiat pair. The size of the dot is inversely related to the variance of these two measures, with larger dots suggesting more stable dislocation and depth.

Scanning this plot makes it easy to see which corridors are limiting factors. The above plot highlights Bitso and Bitstamp, which are important exchanges for the USD-MXN corridor of ODL. You can see Bitstamp has twice the liquidity as Bitso up to the same depth, and both have relatively small dislocation.

By animating this plot over time, we can see whether liquidity is improving as ODL transaction volume increases. Ideally, exchange corridors move closer to the center line and further to the right, as dislocation disappears and liquidity grows.

Next Steps

Between this visualization and the implied FX metric, we have a good understanding of the liquidity at an exchange at a snapshot in time. The implied FX rate lets us compare ODL with traditional FX rates, while the dislocation vs. depth plot lets us compare the price competitiveness and capacity across exchanges and currencies.  

Yet, both of these metrics are static, calculated by sampling the order book across exchanges at fixed points in time. Liquidity is dynamic. Future work will involve modeling liquidity dynamics and resiliency to inform ODL payment execution strategies.

If you’re interested in working on emerging blockchain and cryptocurrency technologies, we’re hiring!

Photo by Chris Liverani on Unsplash