IRS approval and the consequent announcement of a distribution date (oc-
curring on the same day) cause the stub to go from negative to positive in
two cases, UBID and Palm. Thus in these cases the market is acting as
though there is significant news on these days.^4
In one case, Xpedior, the distribution never occurred. On March 22,
2000, the parent company Metamor announced that another firm was ac-
quiring it. Xpedior’s stub rises markedly on this day. However, one could
argue that on and after this date, Xpedior’s stub has little meaning since the
distribution is presumably canceled. On the announcement day, although
not explicitly canceling the spin-off, the acquirer failed to confirm the spin-
off and instead announced that it had gained control of Xpedior and was
investing additional money in it.
The picture is one of predictable idiosyncratic movement in stubs. Stubs
start off negative and then get positive. This pattern is repeated over time
and does not appear to reflect systematic exposure to some common factor,
but rather idiosyncratic developments.
We draw two conclusions from the analysis so far. First, we are able to
identify six cases of clearly negative stubs. We do not think that the pro-
portion of negative stubs, one-third of the cases we study, is particularly
significant. As we stressed above, a negative stub indicates a gross case of
mispricing. Even a single case would raise important questions about market
MISPRICING IN TECH STOCK CARVE-OUTS 141
! "#$
%&' # #
"'(
# # "' ) & "#
# ""
& $' #Figure 4.3. 3Com/Palm stub, March 2, 2000–September 18, 2000.
(^4) In one case, Retek, the stub has less of a trend. Because of the uncertainty about the ulti-
mate distribution ratio, Retek’s true stub is not totally clear in July 2000. The reaction of the
stub to the distribution announcement has a different meaning for Retek since the announce-
ment contains important quantitative information. The announced ratio was 1.24, whereas 10
days earlier an analyst report contains an estimate of 1.40.