Penalization Scheme for Commitment incentivization

[fredo]

Until an enforceable commitment is given and the upcoming trade is final the free option will fundamentally be existent. Another way to incentivize traders to go with the commitment even if a free option for a trader is there is introducing a penalization scheme which works contrary to one's free option. The economic incentive is moved towards being honest. In fact, the goal must be to always have the highest payout on being honest.

Specification

Third Party

In this scheme there is a third party judging the participants based on their statements. The third party has no direct insight into the network so therefore it does not no the actual outcome of the event. Penalties will be given based on the statements of both parties.

Proving the opponent's behavior

Due to the nature of direct messaging, one can never prove absolutely the arrival of his own message to the opponent. Therefore participants can only provide proof of the opponent's correct behavior.

There are three possible decisions to make based on the action of the opponent

• The opponent sent the payment (correct behavior)
  -- Provide the Proof (YES)
  -- Deny the receipt of the payment (NO)
• The opponent did not send the payment (misbehavior)
  -- Deny the receipt of the payment (NO)

Upon receiving NO the third party cannot define the misbehaving party.

Decision Matrices

Scenario 1: No initiated payment

In this scenario, none of the participants initiated a payment or better said none received a payment from the other.
Since you have to provide the signed message of the opponent to claim a YES it is not possible to forge a receipt.
The outcome will be: Both participants state NO

Scenario 2: Exactly one participant does not receive payment

In this scenario, exactly one of the participants will receive a payment. This means that most likely one party was not sending a payment. Let's call this participant Alice. The opponent's answer, Bob's answer, will, therefore, be NO.
Alice knows Bob's answer already. She can make her decision based on this fact. Alice still has the possibility of saying YES or NO.

Scenario 3: Both participants receive payments

In this scenario, both participants started to behave correctly. Both sent their promised payments. This is the only scenario where the exchange can be claimed as successful. Both parties can testify the other's correct behavior and make the exchange officially successful. It should always be in the best interest of any participant to give its YES.

Nevertheless, participants can have the option to say NO.
Disagreement, one person stating NO and the other stating YES means that one party is lying over the other. In this case, the party saying NO is likely to be the betraying party. At least this is the case for scenario 3. In should thus be penalized stronger than the other party.

If in the last case, both parties are stating NO the third party will assume that the trade have not went through.

Payout Matrix

Assuming that the volatility over the commitment time is lower than 5

(A/B)	B_YES	B_NO
A_YES	(10/10)	(4/2)
A_NO	(2/4)	(5/5)

Iterative decision matrix

We assume that on a second layer exchange participants tend to be involved in more than one trade because establishing connecting channels is expensive and only pays back after a certain amount of trades. Therefore multiple trades between participants or of participants is likely to happen. There exist iterative strategies which can be used by the participants to get the best payout based on the opponent's past decisions.

Behavioral Payout

After a certain period of time deposits of misbehaving parties have been slashed continousely. To incentivize the participants to engage in exchanges a good behaving participant should be paid out proportionally. Good behaving parties receive their share of the seized deposits.
Ideally this amount should diverge against 0 thus creating a good behaving environment. Behaving correctly in an misbehaving environment should instead bring a high rebate.