220 S. Drucker and M. Puri
and firms that are performing poorly.^29 Second, the authors find that firms that have
their loans sold are not the worst performers in their respective industries during the
year before their loan was sold, indicating that ex-ante, publicly available information
alone may not have allowed outsiders to identify the true weakness of these firms. These
results support the view that loan sales by the original lender provide negative private
information to outside investors.
6.2. Non underwriter-bank loans and public security pricing
In addition to conveying private information to the market through announcements and
sales, the analysis inFama (1985)andDiamond (1991)indicates that bank loans reduce
the need for outsiders to generate duplicate information. This may allow bank lending
to reduce overall information costs. Further, bank loans can help resolve information
asymmetries between management and outside investors that could induce managers to
refrain from issuing equity and foregoing positive net present value investments (Myers
and Majluf, 1984). One way to test if bank loans reduce information-related costs is
to empirically examine the impact of existing loans on the pricing of a firm’s public
security issuance. If bank loans reduce information costs, then the existence of bank
loans should result in higher security pricing.
James and Weir (1990)investigate how an established relationship with a bank af-
fects IPO underpricing. The authors develop a theoretical model that predicts that due
to the information benefits of having an existing lending relationship, firms with an es-
tablished lending relationship will experience less severe underpricing when they go
public. To test the model, the authors collect a sample of 549 IPOs for non-financial
firms that occurred between 1980 and 1983 and identify 417 firms with existing bank
borrowing relationships. After controlling for factors that have been identified to affect
underpricing (i.e. the reputation of the underwriter, the age of the firm, the offering
size, and the shares offered by insiders), having bank loans outstanding or a bank-credit
agreement significantly reduces underpricing by 8.5 percent. This result suggests that
bank loans reduce the information costs associated with issuing public securities.^30
In addition to having a positive effect on equity issuances, existing lending relation-
ships may also reduce the costs of public debt financings.Datta, Iskandar-Datta, and
Patel (1999)examine this possibility. The authors argue that if banks have superior
monitoring ability, then the presence of a bank lending relationship should lower infor-
mation costs associated with raising public debt, which will be reflected through a lower
(^29) For this test, the authors are able to expand their sample to 53 firms that have a sub-par loan sold. They
were forced to use the smaller sample of 15 loan sales in the event-study due to missing information on the
precise date of sale.
(^30) However, the authors do not find a statistically significant difference between the effect of bank loans and
long term debt on IPO underpricing, which does not support the hypothesis that bank loans play a special role
in reducing information costs for IPOs.