Will Manifold offer a perpetual swap this year?

Basic

Ṁ4910

resolved Jan 3

Resolved

ALL

Perpetual swaps are financial wizardry that let people bet on the value of any asset. Manifold could eg. offer a SOL/M$ perpetual swap contract, tracking some known SOL oracle - and then people could have their portfolio track the price of SOL. See https://medium.com/derivadex/what-are-perpetual-swaps-130236587df2 It's a much more elegant way of phrasing almost all the questions on eg "Will TSLA close above $710", because now you can directly invest in TSLA instead. The demand for this is proven by things like https://manifold.markets/tag/economics Other usages: - Perp swaps on eg the portfolio value of a particular Manifold trader would allow you to directly bet on how that trader is performing - Perp swaps on Elon Musk's tweet count would allow you to speculate on Elon tweeting more or less often than he does today Questions: - Is there a mathematical equivalence between perp swaps and numeric markets? - Or with binary CPMMs, perhaps continuously resolved?

This question is managed and resolved by Manifold.

Get

1,000

and

3.00

12 Comments

15 Holders

210 Trades

Sort by:

predicted NO

predicted YES

https://research.paradigm.xyz/cartoon-guide-to-perps

predicted YES

I think reversable partial resolution of shares Ala https://manifold.markets/MartinRandall/will-partial-resolution-of-yesno-ma would actually let market creators do the "longs pay shorts" side of a perp swap. Aka while 60% is the true answer, the creator allocates M$ 0.6 to each YES holder and the rest to NO; then if it drops to 30% the creator the creator moves M$0.3 from each YES holder to each NO holder. Maybe this would fully work! The last bit of a perp swap is the collateral, which I believe is just the M$ put in the beginning. Still need to think through if & how leverage plays into this.

I think this is a cool idea for the future. Perps let you bet not only on any asset, but any time-series (e.g. global temperature indices, my weight over time, politician approval polls, etc.). Still betting no though, because executing this feature well would be difficult and potentially distracting from more important features.

Are perpetual swaps common in traditional finance? If not, why? Potential starting point for answering that: https://en.wikipedia.org/wiki/Perpetual_futures

@dreev This is a very interesting question! Perps are an extremely niche asset on Wall St today. My guesses for why this is so (in decreasing order of importance): 1. Inertia / path dependence: Futures contracts were invented before computers, and perpetual rolling would have been too difficult from a back-office perspective for the 19th century CME. 2. Switching costs: Wall St is already equipped to deal with quarterly/monthly rolling futures contracts. A perpetual S&P future might be a little easier to trade/clear/settle, but not that much, and switching to a new system is a big cost. 3. Institutional capture: Trading firms actually benefit from making market microstructure as convoluted as possible because they can invest the resources to exploit inefficiencies in the long term in a way which smaller, shorter-term market participants cannot. 4. Regulation: Getting any new type of derivative approved by regulators is a pain.

predicted YES

@SG That makes a ton of sense. Thanks! I think sometimes the answer is "no because tradfi tried that once and it was a disaster" and defi insists on re-learning such lessons the hard way. :) But in this case your answer is pretty convincing that that's not the case for perpetual swaps.

predicted NO

Once you allow mana transfers someone can sell mana on the secondary market and buy perpetual swaps or any other financial gadget. Not seeing the purpose.

predicted YES

Haha I think there's a lot of value in having everything be on a single interoperable chain. It's a lot of extra friction to move out your mana in that way.

I like the idea. It is a generalisation of the current markets to allow trading on anything. I guess the value would need to have an upper limit unlike stocks / assets.

This could be interesting and I like the idea. But to respond to this part in particular: > It's a much more elegant way of phrasing almost all the questions on eg "Will TSLA close above $710" I think people are making and trading on these markets not because they want to directly invest in the securities, but because they want to make predictions, and these are an abundant source of clean, easy questions that resolve quickly. I don't really like them personally, because I think the financial markets already are aggregating far more dollars into expressing views on roughly equivalent questions. A "Will X close above $Y?" question is roughly equivalent to a bull call spread, as I described here: https://manifold.markets/EnopoletusHarding/will-the-sp-500-be-higher-on-june-1. One interesting thing though is that "Will X be above Y" questions are the most natural (and simple and cheap) to express in a prediction market, while they are hard and relatively expensive to express in the financial markets. To implement it by actually trading options is not great because it requires taking a fairly large position and the liquidity is typically poor (high bid-ask spreads and low volumes). So I think there is value of the "Will X close above $Y" market is in making those questions simpler and more liquid to trade on. Though I think that a binary "Will X close above Y" question just isn't usually that interesting to the financial markets, because you usually also care about things like how much above or below. Overall though, supporting more financial engineering is potentially interesting because it opens the doors for people to develop more interesting markets on top of these structures.

@jack Ah, that's a good point! One very cool thing about having perpetual swaps implemented is that you can actually then treat the Manifold perp swap on TSLA as an "on-chain" oracle for your "Will TSLA close above 710" question. So there's no more resolution risk or delay! Of course, we could already do this by letting people plug resolutions into a data/API thing (which, I actually think we ought to do eventually). But having it all be in the Manifold system seems much neater

Related questions

Related questions