In a blog post published on March 8, Hayes stated that stablecoins do not necessarily need to be decentralized, and that "overcollateralised stablecoins such as MakerDAO's DAI and algorithmic stablecoins such as TerraUSD are fundamentally unnecessary (sic)". He argues that the main issue with stablecoins is not centralization, but that no reputable and established banking institution is willing to launch its own.
Hayes believes that if major banks like JPMorgan were to launch their stablecoin, it could potentially harm the business of other banking partners and dampen the company's future earnings. However, the bank would benefit from additional deposits and could lend those deposits and earn interest on them with no risk to the Federal Reserve.
"The problem with stablecoins is not centralization but rather the lack of involvement from established banking institutions," - Hayes said.
As an alternative to existing stablecoins, Hayes proposes a NakaDollar which he describes as a 'synthetic USD'. The solution is basically a bitcoin with hedging against BTC/USD exchange rate volatility.

To create 1 NUSD, one must deposit $1 worth of Bitcoin on a derivatives exchange and short 1 Bitcoin with a Bitcoin/USD inverse perpetual swap. The perpetual swap's payoff function will compensate for the loss of the value of the original $1 of Bitcoin when Bitcoin drops against USD.
Hayes envisions that a close circle of 'member' exchanges that list liquid inverse perpetual swaps for NUSD. The author accepts that it would not be decentralised and the points of failure in the NakaDollar solution would be centralised crypto derivatives exchanges. Further, he suggests that only a few firms or individuals, the 'authorised participants' would be allowed to create and redeem NUSD directly from the DAO.
The main risk of perpetual swaps is that they source their benchmark price for external oracle feeds. To mitigate this, Hayes proposes to source the price from the member exchanges, weighted by their participation share.
The other risks that Hayes outlines in his articles are 1) the loss of Bitcoin by a member exchange due to internal theft or hack 2) negative funding when market sentiment makes short swap holders to pay interest to the longs, and 3) a socialized loss, when the shorts' profits are so large that exchanges cancel them. Hayes proposes to establish an emergency 'sinking pool' as an answer to all three risks.
Hayes proposes that to fundraise for the sinking pool and enable holders to vote on operational issues, a decentralized autonomous organization (DAO) with its governance token, NAKA, be established. The funding generated from holding the perpetual swap can be routed back to the DAO. Hayes further noted that both NUSD and NAKA governance tokens would be ERC-20 tokens on the Ethereum blockchain.
The idea sounds intriguing and, if the risk of dependence on the 'member exchanges' and 'authorized participants' is properly governed with an effective DAO, it can be the shortest link between Bitcoin, the ultimate decentralization and stablecoins, the ultimate crypto use case. We look forward to Observing it.