Information Transmission between Banks and the Market for Corporate Control

How do economic incentives govern the diffusion of private information and resource allocation in financial markets? Attempts to microfound this relationship are centered on the incentives to produce and share information in social networks (Herskovic, 2020, and Leister, 2021), such as those of interconnected banks. However, empirical evidence is limited to the economic consequences of social connections and the extent to which they facilitate social learning and the transmission of private information (Kuchler, 2018). While banks’ key economic role is typically seen in collecting, processing, and producing private information relevant for financial decisions (Morrison, 2007), little is known about banks’ incentives to disseminate private information within their relationship networks and how this affects market outcomes.

In Bittner et al. (2022), we study under what circumstances and how incentives matter for the transmission of private information between banks. We use syndicated-loan networks of banks in conjunction with administrative security-transaction data to infer information flows around the announcement of corporate takeovers. Information regarding imminent takeovers may spill over from banks that serve as advisors in the market for corporate control to other banks when both groups are also active in the syndicated-loan market. This context helps to identify the source of private information. We make use of the fact that the incentives in passing on information about upcoming transactions vary across target and acquirer stocks as well as advisors.

Our empirical strategy is based on the idea that banks establish (social) ties when engaging jointly in the syndication of loans, and that some of these banks simultaneously act as advisors to target and acquirer firms in M&A transactions. Stronger ties facilitate the exchange of M&A related information between bank pairs. To measure the strength of banks’ ties to target and acquirer advisors, we use the fraction of jointly issued syndicated loans. In doing so, we can contrast the relative importance of the advisor to the bank, whose proprietary trading we observe, and vice versa.

Note: The figure shows the point estimates and 95% confidence intervals for 9 days prior to the announcement and 5 days afterwards, based on the following regression specification: Return (cumulative)_st = β_t Σ⁵_{t =-9} Takeover_st+ δ_t + γ_s + ε_st, on a sample at the security-date level from 30 days prior to 5 days after the announcement. Standard errors are double-clustered at the security and date level.

In particular, we can keep constant private information about the imminent merger while exploiting the fact that incentives for leaking this information vary across traded stocks and advisors. This is because announcement returns are positive primarily for target, rather than acquirer, stocks. As can be seen in Figure 1 (left), target stocks have highly economically and statistically significant announcement returns, controlling for security and date fixed effects, whereas this is not the case to the same extent for acquirer stocks (Figure 1, right). This suggests that trading on private information about imminent takeovers is profitable primarily in target stocks, i.e., by purchasing target stocks ahead of announcements. The latter is, in turn, reflected in a more emphasized run-up in targets’ stock prices ahead of takeover announcements.

If traders—e.g., other banks—act on this information and buy target stocks prior to takeover announcements, the takeover price increases, which implies that the target shareholders receive a larger share of the surplus. This would, however, not be in the interest of the acquirer advisor. In contrast, the incentives of privately informed traders and target shareholders, which are represented by the target advisor, are aligned.

We present evidence of the trading behavior of banks that vary in the degree to which they are connected to the advisor of the target firm in a given takeover. Figure 2 shows that connected traders buy target stocks ahead of takeover announcements, potentially reflecting that they take advantage of private information they have accrued through the target advisor.

Note: Trading by connected banks refers to traders having joint syndicated-lending activity with at least one of the target advisors one year prior to the M&A announcement (solid blue line). Trading by non-connected banks is shown by the dashed red line.

Using administrative data at the bank-security-date level from Germany, we can estimate the effect of banks’ connectedness to target and acquirer advisors on their trades around international takeover announcements. Our most important explanatory variable of interest measures the intensity of a connection between a trading bank and a given deal’s M&A advisor, namely by the number of joint syndicated loans scaled by the total number of syndicated loans granted by the advisor or the trading bank. As such, our measure captures the relative importance of the trading bank for the advisor’s syndicated-loan business. The granularity of our data allows us to control for time-varying unobserved heterogeneity at the security and the (trading) bank level. Doing so, we find that banks closely connected to the target advisor purchase more shares of the target, but not of the acquirer, in the 30 days prior to the takeover announcement.

A trading bank that is more connected to the target advisor by one standard deviation purchases, on average, 54% more of the target stocks in the 30 days prior to the announcement. This finding lends support to the view that banks that are more important for the target advisor’s syndicated-loan business are more likely to obtain private information about the imminent announcement of the takeover bid permitting the connected bank to buy target stocks ahead of the announcement at relatively low prices and realize trading profits due to the positive announcement return. These effects are stronger when the potential trading gains are larger, i.e., for higher announcement returns, deals that are completed in a shorter amount of time, and for cash, as opposed to stock, transactions.

When banks that are more connected to target advisors purchase target shares ahead of takeover announcements, they do not merely emulate advisors’ trading behavior, as we do not find advisors to act on their private information and purchase target shares themselves. This suggests that target advisors leak private information about imminent takeovers. At the deal level, we then show that they benefit from leaking such information to connected traders as it helps to drive up the pre-announcement stock price of the target and, as such, the final offer price. This does not come at the cost of lowered deal success probabilities, which would diminish the expected revenues accruing to the target advisor. Our evidence therefore suggests that target advisors have an incentive to leak this private information, and they share it effectively with connected banks that actively trade shares of non-financial corporations. By affecting the offer premium, this has real implications for the division of surplus in M&A transactions, without any repercussions for the reputation of the target advisor. On the contrary, our findings are consistent with a positive feedback effect for target advisors that successfully represent target shareholders’ interests.

In Bittner et al. (2022), we provide evidence that M&A advisors share private information about imminent takeovers to closely connected banks, and that they do so in an incentive-compatible fashion. We uncover these connections using the network of banks in the international syndicated-loan market. Only target, rather than acquirer, advisors share the information with connected banks that purchase additional target stocks before the announcement and, as such, at lower prices. These effects are more emphasized when takeover announcements are associated with higher announcement returns, which is the case when deals are completed faster or are in cash. The additional pre-announcement demand drives up the pre-announcement price and thereby contributes to a higher offer premium without diminishing the probability of a successful takeover bid. Information leakage thus benefits target shareholders and ultimately the target advisor, reflecting the idea that bank networks aid the establishment of mutually beneficial relations.