Ch. 5: Banks in Capital Markets 225
proportion of loans than an unconstrained bank (Haubrich and Thomson, 1996; Pavel
and Phillis, 1987), unconstrained banks are more likely to buy loans (Demsetz, 2000),
and strong local origination opportunities are positively related to loan selling (Demsetz,
1994, 2000).
Recent papers have explored the effect of loan sales on corporate borrowers. It may be
costly for borrowers to have their loans sold, particularly if they need to renegotiate their
loans in the future, as they will have to deal with additional lenders that may not take a
long-term view of the company’s prospects. Further, there is a concern that loan sales
harm lending relationships.Guner (2006)examines if borrowers receive an offsetting
benefit through lower loan interest rates.Guner (2006)identifies banks that were active
loan sellers during the 1987 through 1993 period and shows that borrowers of these
banks indeed received lower loan interest rates. Importantly, the interest rate reductions
are concentrated among borrowers that are more likely to have their loans sold based on
ex ante characteristics.Drucker and Puri (2006), using data that covers the time period
1999 through 2004, show that borrowers whose loans are sold receive additional bank
loans, both in the year of the loan sale and in the future. These results are consistent
with loan selling increasing borrowers’ access to bank loans. Contrary to concerns that
lending relationships are harmed by loan selling,Drucker and Puri (2006)show that
borrowers whose loans are sold are more likely to retain their lending relationships.
One explanation is that loan sales let banks manage their lending risks up front, which
permits banks to extend loans to their relationship borrowers in the future.
7.4. Bank organizational form
There is a growing amount of work on the nature of information collected by banks
from their clients, and on how the organizational form of the bank may be more con-
ducive to collecting some kinds of information as opposed to others. A key empirical
finding is that large banks tend to lend to large companies and small banks tend to lend
to small companies (seeBerger, Kashyap, and Scalise, 1995; Berger et al., 1998; Berger
and Udell, 1996; Nakamura, 1994; Peek and Rosengren, 1996; Strahan and Weston,
1996, 1998; Sapienza, 2002). Stein (2002)argues that the key difference between small
and large business lending is that small business lending relies on “soft” information,
which is information that cannot be directly verified by anyone other than the agent
who produces the information. Small banks are better at processing soft information
while large banks are better at processing verifiable “hard” information, such as finan-
cial statements, public credit ratings, and formalized records.
Since research shows that relationships are important for small companies (see e.g.
Petersen and Rajan, 1994), it is vital to understand the effects on small firms of the
growth in the size of banks and the increased reliance on hard information. There are
a few empirical papers that examine the role of hard and soft information in the credit
decisions of banks.Liberti (2002)examines a hierarchical structure change in a corpo-
rate commercial lending division of a foreign bank in Argentina. He finds that managers
with more independence base their pricing decision more heavily on soft information