Chapter Ⅱ .The Wealth Effects of Oil-related Name Changes on Stock Prices:
4. Empirical results
4.1 Wealth effects of corporate name changes involving oil-related
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
4. Empirical results
4.1. Wealth effects of corporate name changes involving oil-related terms
In estimating abnormal returns around the event day, I exclude firms that have no trading volume from day -1 to day +1 relative to the event day to avoid the impact of stale trading. I calculated abnormal returns for seven event windows. As illustrated in Figure 2.1, there are more companies adding “oil” or “petroleum” to their names than deleting them during the hot market. If investors are affected by cosmetic changes in corporate names and managers take advantage of investors’
irrationality, I expect wealth effects of corporate name changes would be varying across different market conditions. Panels A, B, C, and D of Table 2.3 report the abnormal returns during the full sampling period, normal market, hot market, and the financial crisis period. For each time period, I present the results for the combined sample, U.S. sample, and the Canadian sample. .
[Insert Table 2.3 here]
Panel A of Table 2.3 shows no evidence of strong market reaction for the full sampling period. None of the combined sample, the U.S. sample, and the Canadian sample illustrates a significant abnormal return on the event date. These results on the event day are aligned with previous results in Howe (1982), Horsky and Swyngedouw (1987), Bosch and Hirschey (1989)6 and Karpoff and Rankine (1994), but are inconsistent with those in Lee (2001), Cooper, Dimitrov, and Rau (2001) and Copper et al. (2005). Both the combined sample and the U.S. sample have a significant and negative CARs in the (+1, +120) window. Most of the CARs for the Canadian sample
6 Bosch and Hirschey (1989) find a positive preannouncement effect followed by a negative post-announcement drift.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
are negative, but only the one in the (+1, +30) window is statistically significant. The paired-test indicates that the CARs for the Canadian market are significantly lower than those for the U.S. market in 3 event windows.
As shown in Panel B of Table 2.3, when oil prices were relative stable, corporate name change failed to induce significant market reaction. There are no significant CARs across all event windows and all samples during the normal market period.
Panel C of Table 2.3 tells a totally different story for the hot market period.
There is strong evidence that investors respond positively to corporate name changes involving oil-related words when oil price was surging and the recent financial crisis has not been exploded. For the combined sample and the U.S. sample, the abnormal return on the event date is significantly positive (6.97% and 7.92% respectively). The positive wealth effect carry over to post-announcement periods; CARs for the combined sample are significantly greater than zero in the (+1, +60) and (-15, +60) windows. The U.S sample illustrates an even stronger positive reaction. The CARs for the U.S. sample are significantly positive in six out of seven event windows. The CARs in the pre-event window (-1, -15) for both the combined sample and the U.S.
sample are also positively significant, this indicates a possible leakage of information before the announcement date. The finding for the combined sample and the U.S.
sample during the hot market period is consistent with the investor mania hypothesis.
The results for the Canadian market are similar to those during the normal market period. None of the CARs for the Canadian sample is significantly positive; a significant and negative CAR is found for the post event window (+1, +120). This result is contrary to the conjecture of investor irrationality during an abnormal market condition. Canadian investors seem not to respond to corporate name change involving oil-term names regardless of the market condition. The pair-t tests show
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
that CARs for the Canadian market are significantly lower than those in the U.S.
market in five event windows. It is evident that U.S. investors and Canadian investors react to corporate name change involving oil-related terms differently.
The above results illustrate that investors react positively to corporate name changes when oil price surge. However, when the financial crisis exploded, the anxiety of overall financial and economic stability may out weight the positive reaction to corporate name change. Even oil price was continued to surge between July 2007 and June 2008, positive market responses are vanished when the period (from July 2007 to June 2008) is added to hot market period; this highlights the adverse impacts of financial crisis.
To further assess the impact of financial crisis on the wealth effects of corporate name changes, I examine the market reaction during the crisis period (from July 2007 to June 2009). The results are shown in Panel D of Table 2.3.
Panel D show negative and significant CARs for relative long event windows for the combined sample and the U.S. sample. CARs are significantly negative in four event windows (+1, +60), (+1, +120), (-15, 60), and (-15, +120). There are only five name changes for the Canadian market, it is not large enough to draw meaningful conclusion. As oil price continued surging between July 2007 and July 2008, the negative market reaction found during the financial crisis period implies that overall market sentiment dominates the market condition in the oil industry.
Together with the results for the normal and hot oil market periods, I conclude that while oil price surge, investors respond positively to corporate name changes, however, the positive reaction vanished as the recent financial crisis exploded.
Increasing volatility and uncertainty wipe out the optimisms generated by surging oil prices. It is convincible that investors expect economic hardship caused by the
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
financial crisis, which in turn, would hurt the oil industry and high oil price is not expected to be sustainable.
Figures 2.4 and 2.5 compare the CARs for the U.S. and Canadian firms during the hot market period and the financial crisis period, respectively, In Figures 2.4, the CARs for the U.S. sample continue rising after the announcement day. On the other hand, the CARs for the Canadian sample are below zero over a large portion of the event window.
As shown in Figure 2.5, during the financial crisis period, the CARs for the U.S.
sample are slightly above zero within a short window around the event date, and then show a significant downward drift. For the Canadian sample, the CARs never rise above zero and the downward trend is more acute than that for the U.S. sample.
[Insert Figures 2.2, 2.4 and 2.5 here]
4.1.1 Addition versus deletion of oil-related terms
To further explore the impact of corporate name change events, I divided the sample into addition and deletion of oil-related terms and the results for various sub-periods are reported in Table 2.4. Panel A of Table 2.4 shows that companies adding the word “oil” or “petroleum” to their names show significant CARs of 13.5% and 36.7% in event windows (-15, -1) and (-15, +60) respectively for the combined sample. In contrast, CARs of the deletion sample are negative in four event windows.
Adding oil-related term seems to provide a positive signal while a deletion conveys negative information.
Market conditions have a great impact on the results. The significant and positive CARs for firms adding oil-related terms exist only during the hot market, in
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
which, CARs are significant in four event windows including the event date. None of the CARs during the normal market and the crisis period are significantly different from zero. Similarly, the negative market responses related to deletion of oil-related terms occur only during the crisis period. The CARs during the crisis period are significant negative in five event windows.
As for the U.S. firms, Panel B of Table 2.4 shows that adding “oil” or
“petroleum” to company names have more positive wealth effects than the combined sample. Positive CARs are evident during all sampling periods, except for the financial crisis period. Strongest positive market reactions are found during the hot market period, in which, CARs are significantly greater than zero in five event windows. During the normal market, U.S. firms adding oil-related terms to their name also show positive market reaction within 2 months after the event date. The results for firms deleting oil-related terms are parallel to those for the combined sample. A deletion of oil-related term is associated with significant and negative market reaction during the full sampling period and the financial crisis period. As more than 71% of the 35 U.S. name changes during the hot market period add oil-related terms, these results are consistent with the conjecture that managers try to benefit from investor sentiment by adding oil-related terms to their corporate names when the oil market is hot.
The Canadian sample shows no evident of significant market reaction regardless of addition of deletion of oil-related term. As seen in Panel C of Table 2.4, for the full sampling period, Canadian firms adding oil-related term has a marginally significant and positive abnormal return on the event day, but it quickly reverts and shows a significant CAR in the (+1, +30) window. Most of the CARs for other periods are insignificant. Similar results are found for the deletion cases.
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Results in Table 2.4 suggest that the U.S. market treats an addition of “oil”
or ”petroleum” to corporate names as a major signal for post-announcement windows, while a deletion of “oil” or ”petroleum” from company names conveys less important information. However the positive effect exists only during hot market period.
Negative market reactions to deletion of oil-related term are intensified during the financial crisis.
[Insert Table 2.4 here]
4.1.2 Major versus minor name changes
In Table 2.5, I provide results for major and minor name changes. A major name change refers to the case that the new corporate name is entirely different from the original name. A minor name change allows investors identify the association with the original name.
As Panel A of Table 2.5 shows for the combined sample, major name changes yield positive wealth effects only during the hot market period. In contrast, the CARs of major name changes are significantly negative in four event windows during the financial crisis period. Investors do not respond to minor name changes across different time periods. The significant CARs during the hot market and the crisis period support the claim that major name changes convey stronger information.
However, a major name change was received as a positive signal when oil price surged, but a negative signal when the financial crisis exploded. This further indicates that the overall market sentiment caused by the financial crisis plays a dominate role in determining investors response to corporate name changes. The U.S. sample shows the same pattern of CARs as the combined sample during the hot market period and
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
the financial crisis. For the full sampling period, positive abnormal returns exist before the event date but are following by a negative drift. Panel C shows that during the full sampling period, Canadian firms announcing a major name change has a significant and positive abnormal return on the event day, but the positive reaction is not sustainable. The relative sample small size makes it hard to draw conclusion for other sampling periods.
[Insert Table 2.5 here]
4.1.3 Oil-related sectors versus oil-unrelated sectors
Results for firms in the oil-related sectors versus firms in oil-unrelated sectors are presented in Table 2.6. In Panel A of Table 2.6, significant and positive abnormal returns are observed for firms in the oil-related sector in the (-15, +60) windows. The price reactions of oil-unrelated firms are weaker; none of the CARs for firms not belonging to oil-related sectors are significantly different from zero
Results for the U.S. sample are similar to those of the combined sample.
Significant and positive CARs are found in the (+1, 30), (+1, +60) and (-15 to +60) windows for oil-related firms. The positive wealth effects for oil-related firms are stronger and more persistent during the hot market period. On the other hand, name changes of oil-unrelated firms lead to negative market reaction. The negative responses are more acute during the crisis period. Results for the U.S. sample are parallel to those of the combined sample (see Panel A). Significant and positive market reaction to name changes of oil-related firms are found only during the full sampling period and the hot market period, with much stronger results during the hot market. Similarly, negative market responses are found during the full sampling
‧
國立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
period and the crisis period, with stronger negative CARs in during the financial crisis.
Panel C further indicate that Canadian investors are not responsive to corporate name changes regardless of the nature of name changes.
[Insert Table 2.6 here]