Municipal bonds are the primary source of financing for local governments, with the value of outstanding municipal bonds reaching $4.1 trillion as of March 2022. Historically, the default rate of these bonds has been very low. This raises the question of why so many municipal bonds have different issuance costs (yield spread), even with similar credit risks. This study proposes a new explanation based on the social connection between local governments and large institutional investors. In particular, we argue that Municipal Bond Mutual Funds (MBMFs), mutual funds that primarily invest in municipal bonds, allocate more capital to counties with stronger social connections, lowering the financing cost for these local governments.
Using Social Connectedness Index (SCI) provided by Facebook, we measure the social connection between a mutual fund’s investment advisor and a municipal bond’s issuer based on their county location. As shown in Figure 1, we find that MBMFs are more likely to hold bonds issued by socially closer municipalities. Note that we control for the physical distance between fund advisor and bond issuer and other confounding factors.
Figure 1. Relationship between Holding and SCI
Note: This figure shows the linear relationship between Log(% Holding) and Log(SCI). The log values are standardized to have a mean of zero and a standard deviation of one.
We also find that these results are focused on a subset of mutual funds and municipal bonds. Specifically, the finding only hold for actively managed funds, but not for passively managed funds whose managers have little discretion on capital allocation decisions. Furthermore, we find that mutual funds under smaller families, possibly with less resources for investment analysis, are more likely to overweight municipal bonds with stronger social connections. Regarding bond characteristics, the effect is stronger for opaque municipal bonds that face greater uncertainty (i.e., revenue bonds and long-term maturity bonds). These findings suggest that investors rely more on social connections when information is scarce.
Based on the positive relationship between social connection and mutual fund ownership, we examine whether the increased mutual fund holdings through social connection reduces the issuance cost of municipal bonds. To test this idea, following Kuchler et al. (2022), we measure each county’s social proximity to the capital, which is computed as the SCI-weighted average of MBMF capital in all counties in the U.S. A county with higher social proximity to the capital has a stronger social connection with other counties where large MBMFs are located and has greater access to capital, which lowers their funding costs.
Figure 2. Social proximity to capital
Note: This figure shows heat maps of U.S. counties’ social proximity to capital.
We find a strong negative relationship between this measure and the county’ yield spread. A one-standard-deviation increase in social proximity to capital is associated with a 7-basis-point decrease in offering spread. This reduction accounts for 16.7% of the average offering spread in our data, which is economically sizable. This finding implies that social connection broadens the investor base of an asset, facilitates risk sharing, and thus decreases yield spreads of municipal bonds. This argument is also supported by the fact that we find no effect for a subset of bank-qualified municipal bonds, which mutual funds rarely hold. In addition, consistent with the holdings analysis above, the impact is stronger for municipal bonds that are more opaque and uncertain.
Furthermore, social proximity to capital also lowers search costs, as proxied by the gross spread. This suggests that the enlarged demand through social connection facilitates underwriters to resell bonds more easily in the secondary market. We also show that our findings are not driven by other potential channels, such as fundamental risks, underwriter effects, or a subset of large counties with national-level awareness.
Our findings are important for a recent study by Babina et al. (2021), who document that the segmentation of the municipal bond market, driven by state tax incentives, increases municipal borrowing costs. In particular, a tax incentive to local investors may induce concentrated local ownership, limiting risk sharing and leading municipal bond yields to be more sensitive to local supply variation and idiosyncratic local risk. Our study suggests that social connection can partially alleviate the cost of limited risk sharing. For example, as Figure 3 illustrates, Cook County in Illinois has a strong social connection with many counties outside Illinois. By expanding potential investor pools to out-of-state investors, social connection thus facilitates risk sharing and augments the value of municipal bonds.
Figure 3. Social Connectedness Index to Cook County, Illinois
Note: This figure shows the heat maps of U.S. counties’ Social Connectedness Index (SCI) in Cook County, Illinois. SCI is provided by Facebook. Darker green counties indicate higher SCI than Cook County.
Our study also contributes to a recently growing literature in social finance (Kuchler and Stroebel, 2021), as emphasized in the recent American Finance Association (AFA) presidential address by Hirshleifer (2020). Most of the existing studies in this area focus on how social connections can influence individuals’ economic behaviors. However, few studies explore whether and how social effects on individual behavior can further extend to aggregate market outcomes. The exception is a study focusing on the stock market by Kuchler et al. (2022), which shows that firms with greater proximity to institutional investors exhibit higher institutional ownership, liquidity, and valuation; the study did not find an effect on stock prices. We document a strong positive relationship between social connection and mutual funds’ municipal bond holdings, which allows the connected municipalities to sell their bonds at higher prices. Lastly, this study contributes to the literature on the investment choices of institutional investors. While studies show that professional investors invest in familiar assets (Coval and Moskowitz, 1999; Bernile et al., 2015; Sialm et al., 2020), most of such findings are attributed to home bias, which may weaken as the importance of physical distance decreases over time (Da et al., 2021). We provide an additional channel based on social connection, through which institutional managers may exhibit familiarity bias to their connected municipalities.
Jinoug Jeung is a Ph.D. candidate in finance at Goizueta Business School, Emory University.
Jaemin Lee is a Ph.D. candidate in finance at the Goizueta Business School, Emory University.
This post is adapted from their “Social Connection and Financing Cost of Municipal Governments” paper, available on SSRN.
