The Nash bargaining solution manipulated by pre-donations is Talmudic (Q2366923)

The author considers the simple property division problem: we have an indivisible piece of property for which two parties have equal claims. This property is worth a money units to one player, and worth \(b\) units to another. Without any loss of generality, one can assume that \(a\leq b\). In this case, it makes sense to give the property to the second player (who values it the most), and require him to compensate the first one for it. This payment \(p\) cannot exceed \(b\), because then the second player will lose more than he gains. For each \(p\) from 0 to \(b\), the resulting utilities of the players are \((p,b- p)\). A natural way to choose \(p\) is to apply Nash's bargaining solution, for which \((u- u_ 0)(v- v_ 0)\to\max\), where \((u_ 0,v_ 0)\) is a status quo point (in our cases, (0,0)). This leads to a solution \((b/2,b/2)\), with \(p= b/2\). Here comes the problem. The second player knows that he will have to pay. So, before the decision is made on how much he has to pay, he voluntarily sends a certain amount \(m\) of money to the first player, and expresses readiness to send the remaining part as soon as the decision is made. This act does not seem to change the problem, but it does change the status quo point, and thus, it changes the Nash's bargaining solution. By properly choosing \(m\), the second player can decrease his total payoff. The author proves that the smallest possible payoff that the second player can get is \(a/2\). Interestingly, the resulting payoff coincides with what Talmud recommends for this situation.