The purpose of this paper is analyzing whether trust and reciprocity are affected by how rich the partner is or how well the partner performed several tasks with real effort. A trust game (TG) experiment is designed with three treatments. First, a baseline Treatment B in which subjects play a finitely repeated TG. Second, in a Treatment H with history, subjects know the partner’s wealth level reached in the past. Third, in a Treatment E with effort the individual endowment with which the TG is played is endogenous and results from the subject’s performance in three different real effort tasks (maths, cognitive and general knowledge related). The data analysis highlights the importance of past wealth levels (Treatment H) as well as endowment heterogeneity (Treatment E), on the actual levels of trust and reciprocity. Specifically, it is observed that the decision of trustors is positively affected by positive past experienced reciprocity. Moreover, trustors are sensitive to how much money the trustee accumulates each round in Treatment H, trusting more the ones that have accumulated less compared to themselves. In contrast with that, it is remarkable in Treatment E that trustors are sensitive to the endowment level of the trustees, trusting more the partners that have got a higher than own endowment, probably considering that a person that performed better in the tasks is a better partner to trust. As far as second players’ behavior, as the amount received from the trustor increases it is less likely that the trustee reciprocates with higher than or with the egalitarian amount. In Treatments H and E, the probability that the trustee reciprocates with higher amount that the one received increases when inequality in endowment/accumulated earnings favors the trustor. Additional results come from analysis of personality archetypes and socio-demographic variables.