Using a newly constructed data set on Israeli Initial Public Offering (IPO)
firms in the 1990s, we study costs and benefits of universal banking. We f
ind that a firm whose equity was underwritten by a bank affiliated underwri
ter, when the same bank was also a large creditor of the firm in the IPO ye
ar, exhibits significantly better than average post-issue accounting perfor
mance, but that its stock performance during the first year following the I
PO is considerably lower than average. When an investment fund managed by t
he same bank is heavily involved in the IPO as buyer of the newly issued eq
uity, the stock performance during the first year following the IPO is even
lower. This, together with negative first day returns, is indication of IP
O overpricing. We interpret these findings as evidence that universal banks
: use their superior information regarding client firms to Boat the stock o
f the 'cherries'. not the 'lemons' (as measured by post-issue accounting pe
rformance), but that bank managed funds pay too much for bank underwritten
IPOs, at the expense of the investors in the funds. These results suggest t
hat there is conflict of interest in the combination of bank lending, under
writing, and fund management. (C) 2001 Published by Elsevier Science B.V.