SIP-05: Recalibrating SYN's emissions

Googly

When Synapse was launched, the token was trading below $1 and liquidity mining incentives needed to be high in order to attract TVL to a newly launched project and ensure its success.
Since then, the token price did more than x4 and TVL increased to over $700m, allowing SYN to offer DeFi users the most liquid bridges (in addition to being the fastest ones).

However, $SYN's price appreciated faster than TVL resulting in an average APY on stables of 40%, with some pools being even higher. Additionally, this stated APY solely comes from SYN rewards: including the base rewards / trading fees, we would have been even higher.

Three observations can be made here:
i) Incentives are significantly higher than for similar projects implying that we are overpaying for liquidity. Excluding levered solutions, I could find only one established large stable pool with an APY of 30% or more
ii) We now have enough data to estimate utilisation of the different pools and allocate rewards accordingly (the more volume a pool drives, the higher TVL is justified)
iii) It is better to build an emission buffer that we can redeploy when, inevitably, more pools and chains will open on Synapse

With that in mind I would propose the following emission reductions over the next three weeks:

5% reduction per week (14% over three weeks) in emissions to the Arbitrum ETH pool. This would reduce the APY around 12% assuming a constant TVL & price. The cut in emissions for that pool is significantly less than others given the high volume that pool generates (incl. some large transactions)
10% reduction per week (27% over three weeks) in emisions to all stable pools (excluding Polygon). Assuming a constant TVL & price, this would reduce the APY to around 30% in those pools
20% reduction per week (around 49% over three weeks) in emissions to the Polygon pool over the next three weeks. This will inevitably reduce the TVL in that pool but is driven by utilisation: Polygon generates 3x less volume vs the other stable pools

Importantly, monitoring the utilisation of those pools will be key: if the reduction in emissions cuts the utilisation / volume, the emission reduction program would be then stopped or at least re-evaluated.

Tokens that would have normally been minted to liquidity provider will go towards the treasury in order to fund further expansion of the Synapse ecosystem (e.g. $SYN tokens potentially required for CEX listing)

RamielCapital

Googly

FINAL TLDR

Harmony: unchanged at 30k per week (reduced once ONE rewards go live)
Fantom: From 95500 pw to 78000 pw with a mandate to adjust it if necessary, including increasing emissions back, to keep average TVL above $50m (see rationale in original proposal)
Polygon: -10% per week over three weeks. From 95500 pw to 70000 pw
BSC: -12% per week over three weeks. From 111,475pw to 77,000pw
Avalanche/Arbitrum: -14% per week over three weeks. From 111,475 pw to 70,000 pw for stable pools and 140,000 to 89,000 for eth pool

2% weekly reduction in emissions thereafter (subject to conditions below)

Second proposal:
The original sip5 proposal was put forward by using admin fees as a proxy for pool utilisation. For reference, you can see accrued admin fees here: https://synapse.dorime.org/#get-/api/v1/analytics/fees/admin/-chain-

However, doing further digging in the data, it seems (i) that this is not the best proxy for utilisation as a non-signifiant portion from fees can come from LPs churning (e.g. depositing assets and then withdrawing shortly thereafter), (ii) a portion of the fees are generated by same chain swap and chains where situation differs between chains (on some Synapse is already integrated into DEX aggregators, in others it is not yet), (iii) there have been changes in volume per chain since inception.

In that context, the best proxy for pool utilisation is likely the metapool swap volume. This table has been put together by Chi (for whatever reason I cannot post it here, see discord) based on Blaze's chad tracker shows the current situation when it comes to volume
https://synapse.dorime.org/#get-/api/v1/analytics/volume/metapool/-chain-

Key line here is imo since 09 October (i.e. since the new pools opened to reflect latest trend + also difficult to make a proposal on 1.5 week of volume) but a few observations

Since opening of the new pools, BSC drives the highest volume followed by Arbitrum, Avalanche, Polygon and FTM
While there was much more of a differentiation in volumes in the early days of Synapse, BSC metapool volume as a share of the total has reduced significantly a bit: from 50% higher than Polygon since inception to 25% higher / from 23% higher than Avalanche since inception to 21% higher
While Arbitrum has done slightly higher volumes that Polygon/Avax since the pools opened, the differential is not massive (and the 1.5 week of data gives slightly lower volume for Arbitrum albeit I don't think we should factor this too much).

In light of this, I would suggest that BSC has 10% more emissions than Avalanche/Arbitrum/Polygon (half of the difference in volume recently basically, to reflect also the small downtrend in shares). This gap will be reevaluated depending on changes in emissions: if it continues to close, BSC will then be aligned with other chains.

Fantom is in a bit of a different situation whereby volumes have so far been disapppointing. However, I would suggest to limit a cut to the emissions for two important reasons (alpha leaks)

Synapse has applied to the Fantom incentive program. The size of those rewards will depend on Synapse's TVL BUT given the way they are structured, it is important to keep our TVL above $50m. This means that we may have to "overpay" slightly incentives on Fantom for the next month or so, until we become eligible for Fantom rewards and we can then replace some of the SYN emissions by Fantom emissions, allowing us to reduce $ incentives on that pool more in line with volume
We are currently exploring ways for our bridge to be integrated on other protocol's front-end (which are expected to be the main bridges used by the Fantom community)

In that context, I would suggest to approach tailor the incentives on Fantom (with a small safety buffer). I would expect that we can normally target an APR of 25% (28% APY) before swap fee (ie around 30% APY in total) without having any material TVL outflow as we would still be higher than established protocols (excl. levered solutions). If you target an average $55m TVL and allow for a 10% drop in SYN price (two safety buffer here), this leads to 84k emissions per week. This is 12% lower than what we have right now on the chain. I would therefore suggest we go with that: it is a 4% reduction in emissions per week over three weeks

For the Avalanche/Arbitrum, I would suggest we keep the initial cuts suggest by the first version of SIP ie 10% reduction per week over three weeks. This leads to 81k emissions per week (rouding down to 80k for simplicity). Importantly based on similar paramters than above, it gets you pretty much around $55m of TVL per chain

For Polygon, to get to similar figures than Avalanche/Arbitrum, the cut would need to be less severe given the lower starting point. This gets to around 5% per reduction in emission per week over three weeks.

For BSC, you are targeting 88k emissions (10% higher than Avalanche/Arbitrum) so around 7% per week over three weeks.

Note that the launch of Arbitrum & Fantom also increased weekly emission / inflation as it was not funded by emissions to existing pools, effectively increasing SYN inflation.

The last pool is the ETH/nETH on Arbitrum. APY of that pool has been hoovering around 10-12% for a while which seems to be its equilibrium. This means that any significant cut in emission is expected to reduce the size of that pool. That pool has done significant volume over its life but interestingly enough (i) volume was much higher at inception (eg 20k ETH in one signle day in September) when pool was significantly lower and (ii) it has lumpy volume (expected to be driven by large tx from whales which you may want to accomodate from, to some extent), (iii) had a recent drop but unclear whether htis reflects a new trend (ie more competition out of Arbitrum, including our own stable pool). Until them volume was slightly above 2k ETH per day ie a c.5% utilisation This is around half the daily utilisation of other pools. It is also the pool with the highest emission of SYN (140k emissions). In that context, I think we have room for reduction here, despite the hit in TVL it would entail. I would suggest 10% per week over 3 weeks with weekly update to see if there is impact on volume & impact on TVL. Importantly even if you assume a 1:1 impact (worst case scenario), TVL would stay above $150m. .

For Harmony, emissions should stay as per the proposal to the Harmony team (at least $250k in SYN and $250k in ONE).

Future adjustments to emissions rates will be done based on volume metrics, following the same line of thoughts as per the above, as well as community feedback (implement by the team to avoid frequent votes for small adjustments).

Any reduction in emissions will go towards the Synapse treasury to fund expansion of the protocol and its ecosystem (as well as allowing fore more leeway as new pools are opened).

Amended and final proposal as published in the discord

I have been slightly re-running the numbers and came to small adjustments to the proposal (no changes in relative emissions those are purely volume driven). Reasons for that are
(i) ONE was actually a +30k net increase in emissions in line with the approach taken for previous pool to bootstrap liquidity in the initial stages so while the cut over the next three week may look high, we are getting to a similar emission schedule than SYN a month ago,
(ii) use of a VWAP to run the maths rather than spot (10% price difference... so I would a few hours later have come with a different figure),
(iii) some test tx on large (rare) swaps show we have actually quite a bit of room even from there,
(iv) a look at some of the larger farmers of the protocol and the apy on other stable protocols they provide liquidity to (sometimes as low as 15%) and
(iv) a more dynamic approach whereby we consciously slightly decrease the emissions more to try to reduce reliance on mercenary capital and attract more stable capital but are ready to re-evaluate every week within the program the evolution of TVL with a mandate to amend reductions (or even increase emissions back) if it falls below our targets / at a level which could impact bridge performance (as we would vote vote a mandate for a slightly larger reduction, we need to be ready to go back for example to the original proposition if APY and hence TVL falls for example due to exogenous events such as a large fall in SYN' price)

Harmony: unchanged at 30k per week (reduced once ONE rewards go live)
Fantom: From 95500 pw to 78000 pw with a mandate to adjust it if necessary, including increasing emissions back, to keep average TVL above $50m (see rationale in original proposal)
Polygon: -10% per week over three weeks. From 95500 pw to 70000 pw
BSC: -12% per week over three weeks. From 111,475pw to 77,000pw
Avalanche/Arbitrum: -14% per week over three weeks. From 111,475 pw to 70,000 pw for stable pools and 140,000 to 89,000 for eth pool

It the program is a success and syn's price allows it, reintroduction of a minor weekly reduction in emissions (-2%, for reference the Nerve tokenomics had -10% but we are still in growth mode so no need to rush with a higher ongoing decay) until further adjustments are made through a governance vote