The existing literature has proposed several reasons for income smoothing by banks. Among them, tax incentives, managerial self-interest, concerns about financial fragility triggered by worsening investor perceptions, meeting capital requirements, and avoiding supervisory scrutiny. With the exception of the first, these reasons rely upon the asymmetry of information between bank insiders and outsiders. Hence, reducing the level of asymmetric information is expected to weaken the incentives and leeway of bank-managers for income-smoothing and, thus, to be associated with smaller smoothing. Which is the subject of this research. To test for this, we estimate typical income-smoothing equations in which loan-loss provisions, as share of total loans or total assets, is regressed on several bank-specific and economy-wide variables. The existence and extent of income-smoothing is judged from the sign and size of a bank-specific variable, i.e., earnings - also as share of total loans or total assets. A positive sign of this variable is consistent with income smoothing: higher earnings are associated with higher provisions, and vice-versa. The estimated equation also includes time dummies. The sample consists of 543 US bank holding companies and covers the period 1999 - 2011. Using US banks only controls for time-invariant country-specific variables which have been explored in the literature, such as, regulations, supervisory power, creditor rights, quality of institutions and financial development. The time dummies additionally take into account the effect of time-varying country-specific variables. To account for the degree of asymmetric information, we use a dummy that takes the value 1 for the years for which a mutual fund that is a member of USSIF has invested in the shares of a particular bank. USSIF stands for US Sustainability Investment Forum, US's membership association for socially and environmentally responsible investment professionals, firms, institutions and organizations. USSIF funds base their investment decisions on firms' ESG (Environment, Social, Governance) performance. The USSIF funds compensate for the lack of readily available and reliable ESG data and ratings with an intrusive screening and monitoring process. During this process there is active engagement of the top management of the funds and the firms --here, banks--, with annual and, occasionally, quarterly meetings. Our identifying hypothesis is that this process, which amounts to stronger monitoring and, hence, stronger market discipline, is associated with lower asymmetric information. Thus, we expect that income smoothing will be smaller for the banks in which USSIF funds invest. To test for this, we use the interaction term between this dummy with the earnings variable. The preliminary results indicate that there is income smoothing. Yet, confirming our expectations, the negative coefficient of the interaction term indicates that income smoothing is smaller for the banks in which USSIF funds invest. [ABSTRACT FROM AUTHOR]