3.5.1 Result and evaluation of Hypothesis 1
While the calculations behind the results are shown in Appendix 40, Panel A below presents the average abnormal returns (AARs), the cumulative average abnormal returns (CAARs) as well as the related statistical t-test values. The estimation period is narrowed down to 30 days prior and post the announcement date including several event windows. Hypothesis 1 states that Norwegian divestitures create abnormal returns on the announcement date. The result of this study reveals an abnormal return on day 0 of 2,5%, which is highly statistically significant on a 99,8% confidence level. This shows strong evidence for the impact of the divestment on the announcement day, leading to an acceptance of the hypothesis. AAR on each day prior and post the announcement date, on the other hand, do not show a statistically significance at a 95% level.
An interesting observation is the positive AAR prior to the announcement. In fact, three to one days prior to the announcement day (-3,-1), the CAAR is 1,658% and statistically significant on a 95%
level. One explanation could be information leakages in the market prior to the official
announcement. Keown & Pinkerton (1981), among others, emphasizes this leak of information in
the market, indicating that rumors can cause abnormal returns before the announcement day. The same authors assume this price increase also can be explained by insider trading. In contrast, Jensen and Ruback (1983) argue that some investors may be able to anticipate takeovers and their trading includes this anticipation into the stock prices. Veld and Veld-Merkoulova (2002) presents another angle, indicating that stronger announcement effects are observable by a longer event window, as the information processing takes place not only after the announcement day but also prior to it. This might also explain why the event windows, which will be discussed below, have a higher CAAR than on the announcement day.
Further, Panel B presents CAARs with relative statistical t-test values for different event windows.
By expanding the event window to for example three days (-1,+1), the CAAR increases to 2,7%.
However, since the AAR is slightly negative the day after the announcement (+1), the CAAR rises to 2,9%, when solely looking at the two-days event window, (-1,0). All the values are statistically significant on a 99,8% level, possibly explained by the relatively high CAARs.
The graph in figure 6 illustrates that the highest AARs fluctuate around the announcement date.
The greatest CAAR of 4,8% is obtained in the seven-days event window (-3,+3), statistical
24 days after. The positive market reactions of the divestment announcements symbolize that the securities outperform the market, indicating value creation for the vendor companies and their shareholders. However, by further expanding the event window, the positive effect decreases the CAARs and switches to no statistical significance, as shown in panel B. However, when moving further from the announcement date, other events may occur in the period affecting the stock prices. Additionally, a broader window includes too many normal days and the information from the divestitures might already be incorporated in the stock prices, indicating a semi-efficient market.
In addition, the graph also provides an indication of the performance by the companies before the announcement. The CAAR is negative until the day before the announcement (-1). This
demonstrates that the companies on average are underperforming the market, which is, according to Section 2, one of the motives for a company to restructure. Further, an announcement of
divestments may convey information on the valuation of the parent company and the divested part, decreasing the asymmetric information. This is reflected in the positive CAAR after the event date, as the information regarding the divestment seems to be incorporated in the stock prices. As a result, divestment announcements clearly affect the performance of vendor companies in the short term.
3.5.2 Result and evaluation on Hypothesis 2
Hypothesis 2 tests whether the announcement of the “major” divestments yields an even higher positive stock effect than the total sample. Panel C displays AARs and CAARs with the relative t-test statistical values for the sub-sample mentioned in Section 3.2.2, while the calculations are shown in Appendix 40. On the announcement date, the sub-sample shows a statistically significant AAR of 3,6% on a 99,8% level. This is substantially higher than the AAR of 2,5% in the total
-3%
-2%
-1%
1%
2%
3%
4%
-30 -28 -26 -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Abnormal return on Norwegian divestements total sample
AAR CAAR
sample, presented above. Hypothesis 2 can therefore be confirmed, claiming that a larger divestment value creates stronger positive reactions on the announcement day.
Further, panel D presents the CAARs with the relative t-test values for the sub-sample. Unlike the total sample, the AAR on the day after the announcement (1) is positive, resulting in the event window of (0,+1) to be even higher than day 0 with a CAAR of 3,7%. This is also the case for the event windows of (-1,0) and (0,+3), displaying higher CAAR for the sub-sample relative to the total sample. All these event windows show strong statistical significance on a 99,8% level. It is clear that the effect of larger divestments has a greater impact on the stock prices of the vendor companies compared with the total sample.
Like the total sample, the graph in figure 8 indicates underperformance prior to the announcement day by the companies with “major” divestments. The highest CAAR is in the 4 days event window (0,3). However, from this point, a broader event window leads to lower, non-significant CAAR, indicating that the stock price effects are strongest close to the announcement day. Further, compared to the total sample, the graph of the sub-sample shows a more extraordinary effect of AAR at the announcement day relative to the rest of the estimation period which supports the
announcement day are not showing positive AAR, which might indicate less speculation and rumors of the divestments.