Observations from @alexbosworth’s Tweet on Fees

Nik Bhatia
3 min readJul 20, 2018

--

The most common question I’m asked is if I know what the interest rates of a Lightning node might be. Now I finally have an idea, thanks to Alex Bosworth’s tweet today containing valuable data points:

Andreas posted he is making 4204 fees/week on his node, which he put $325,000 on. (week_fee_sum”: “4204”). My node has $7,500 which is about 40x less, but I’m making almost 2x his weekly fees. (“week_fee_sum”: “7829”). Why do I make more with less? Capital isn’t the only factor.

@alexbosworth

https://twitter.com/alexbosworth/status/1019985943321706496

I organized his information into a table and provided some conversions and calculations given the data provided. The three variables required to calculate interest are present in his tweet, giving us an exciting first opportunity at measuring the on-chain time value of bitcoin. The variable “week_fee_sum” contains income and time, and the principal is given in US Dollars. I assumed a BTC/USD price of $7,400.

Node Rates

Conversions and Calculations

1. Converted the given principal (USD) to satoshis.

2. Divided fees/week (satoshis) by principal (satoshis) to attain the weekly return on each node.

3. Converted the weekly rate into an annualized rate by compounding the weekly rate 52 times.

4. Converted the given fees/week (satoshis) into USD.

5. Divided fees/week (USD) by principal (USD) to double check the previous interest rate calculation; the interest rates matched, as expected.

6. Divided fees/week (satoshis) by 1,008 (blocks/week) to attain the average fees/block as another way to observe the data.

Observations

1. The first and most important observation from the data is that Alex Bosworth is showing an annualized return of 0.40% which is material income, especially in an low global interest rate environment. I am pleasantly surprised and believe this number will turn a lot of eyes.

2. The disparity of returns between Alex’s and Andreas’ nodes is enormous. Alex’s interest rate is 80 times Andreas’s rate! Because we only have two data points, we don’t know if these interest rates are high, low, or average relative to the rest of the network.

3. I believe the integrity of the data he provided, but we weren’t given a way to verify it. I believe a mechanism to verify another node’s stated interest rate is crucial to open capital markets.

4. Notice that he volunteered this information despite it being private and proprietary to his node. This is common in capital markets; for example, a company that discloses its financials to secure funding from investors. Alex is advertising his skill and will likely attract investors seeking a positive interest rate on their bitcoin. This is how on-chain time value calculations can lead to off-chain capital markets.

5. I chose to express fees as satoshis/block as an additional way to think about performance between the two nodes. While the number is informative, it doesn’t account for the principal it took to earn that amount and is therefore not an interest rate.

6. His final sentence, “capital isn’t the only factor,” is meant to point out that skill is a driver of returns. Skilled Lightning node operators will drive competition, and node management techniques will evolve and improve as a result.

Conclusion

A very big thank you to Alex Bosworth for volunteering this incredibly valuable information on the economic performance of his Lightning node. We are in the early stages of calculating time value and risk premiums using Lightning Network’s framework. I’m optimistic that Alex will set a trend by publishing his Lightning interest rate, even though he didn’t express it in those terms. Discussions about LNRR and its role can wait until more research is done on individual Node Rates.

Follow me at https://twitter.com/timevalueofbtc

Follow Alex at https://twitter.com/alexbosworth

--

--

Nik Bhatia
Nik Bhatia

Written by Nik Bhatia

bitcoin-native financial theory, adjunct professor of finance and business economics @USCMarshall

Responses (1)