How money is managed in the stock market is of first-order importance, particularly in the United States, where the plurality of individuals participate in equity markets through delegated portfolio management (i.e., professional portfolio managers). A central player in this process is Morningstar – who provides widely recognized and accepted star ratings for mutual funds. These ratings guide investors on where to allocate their money, with higher-rated funds on average attracting significantly more investor capital. Morningstar’s star ratings are based on a clear and standardized process, which helps investors compare funds easily. However, this transparency also opens the door for mutual fund managers to adjust their strategies to receive higher ratings, boosting their appeal to investors and, in turn, the mutual fund managers’ revenues.

Morningstar assigns its star ratings by grouping most equity funds into nine categories (e.g., Large Growth, Small Value) based on the types of stocks they hold (See the figure below). These categories, or “style boxes,” are updated regularly based on changes in a fund’s holdings. When a fund changes categories—which we refer to as “box jumping”—it is re-evaluated and its historical performance is compared to other funds in its new group. For example, if a fund was operating in the Mid-Growth category for 10 years, and changes to Large-Growth, the fund is immediately compared to other Large-Growth funds based on their three, five, and 10 years of past performance. This re-categorization can immediately and significantly change a fund’s rating, holding its actual performance over time fixed.

Our paper presents evidence consistent with actively-managed mutual funds taking advantage of this flexibility in the system. They adjust their portfolios in ways that prompt Morningstar to reassign them to a new category with poorer-performing funds, making their own performance look better by comparison. As a result, on average these funds receive significantly higher fund ratings, sizable inflows of capital and increased management fees; all despite not delivering better returns for investors.

We first show that box jumping funds are twice as likely to receive rating upgrades compared to downgrades in the month they switch categories. From 2003 to 2022, about 35% of box jumping funds receive rating upgrades, while only about 15% experience downgrades. This pattern only emerges following 2002, when Morningstar begins basing ratings on relative performance within the 9 style boxes. Prior to June 2002, ratings were compared relative to all equity funds regardless of style boxes, making it difficult for funds to achieve rating upgrades through box jumping. The figure below shows the yearly average change in ratings for box jumping funds in the month of box jumping. In the five years leading up to 2002, the average change in ratings for box jumping funds was close to zero, whereas in the five years following 2002, it increased to an average ratings upgrade of about 0.33.

Next, we focus on the intermediate steps that funds take to facilitate favorable box jumping. Specifically, we show that funds adjust their portfolio holdings in a predictable fashion based on their specific incentives to box jump. Funds are more likely to buy stocks that facilitate box jumping when they are near the boundary of a neighboring style box that has poorer performance. Conversely, they adjust their holdings to retreat away from neighboring style boxes when being moved to those boxes would result in a ratings downgrade.

We further show that certain types of funds and managers are more likely to engage in box jumping, while others are less inclined to do so. Fund managers with more experience and those who have previously box jumped with rating upgrades are more inclined to box jump. Additionally, funds that are underperforming relative to their peers based on style box or fund family are more likely to box jump. In contrast, funds that have specific style box-related names (e.g., Fidelity Small Cap Growth Fund) or funds with family peer funds already in the neighboring style boxes are less likely to box jump. These findings highlight that the willingness and the opportunity to box jump play significant roles in determining whether a fund will engage in this action of box jumping.

Funds that successfully box jump benefit from increased capital inflows and higher commensurate management fees. Namely, funds that receive a ratings upgrade from box jumping see statistically identical inflow increases as those funds that receive upgrades without box jumping – suggesting investors do not identify or distinguish them (treating all upgrades identically). Additionally, box jumping funds often charge higher management fees after a ratings upgrade. However, by changing their portfolios to fit a new style box, funds may deviate from their core investment strategies, leading to lower future returns. Moreover, the ratings improvements achieved through box jumping are often temporary, with the average box jumping fund seeing its upgraded ratings revert in about three years. This suggests that while fund managers benefit from higher inflows and fees, investors may not receive the returns they expect.

In sum, we provide novel evidence that mutual funds adjust their portfolios to achieve better fund ratings without improving actual performance. Investors seem to be largely unaware of how these ratings are achieved and allocate money simply based on the face value of ratings. Our work raises important questions about the overreliance on simple ratings and the role of information intermediaries like Morningstar in shaping strategic fund manager behavior, and coupled investor decisions.