Subsequently, several sub-hypotheses formulated based on the literature review were investigated to understand the characteristics and motives that drive the positive abnormal return around the announcement. The hypotheses were analysed through cross-sectional tests and regression analy-sis.
The first sub-hypothesis, H1a, relates to whether spin-offs to generate higher abnormal stock returns than sell-offs. We demonstrated the nominal CAAR values of spin-offs to be higher than those from sell-offs, however, we did not find strong statistical evidence that the difference in CAARs was sig-nificantly different than zero, and thus H1a is weekly accepted.
The argument of divesting is often related to a matter of increasing focus, both in the literature and in practice. However, divesting in nature, is focus increasing as the parent company decrease in size. The rationale of H1b and H1c was that increasing focus would be rewarded by the market, and hence create higher CAARs. Thus, higher CAARs for industry and geographical focus increasing divestments were expected. For the total sample and sell-off sample, the differences in CAARs of industry focusing divestments showed mixed results with no statistical significance. Thus, H1b was weakly rejected for those samples. The CAARS for industry focus increasing spin-offs were higher than non-focus increasing spin-offs. Due to no statistical significance, H1b was weakly accepted for the spin-off sample. The test of geographical focus showed results of the opposite signs, meaning that the non-focus samples generated higher CAARs than the focus sample. The results were sta-tistically significant at various levels for the total and sell-off samples, and insignificant for the spin-off sample. Therefore, H1c was strongly rejected for total and sell-spin-off samples and weakly rejected for the spin-off sample.
Another commonly mentioned motive for divestitures, is the reduction of asymmetric information between management and shareholders, resulting in conglomerate discounts. The literature showed asymmetrical information to be an important motive for completing divestments, particularly for spin-offs. H1d and H1e was formulated using idiosyncratic volatility and Tobin’s Q as proxy variable for information asymmetry. By dividing the sub samples into two groups, high and low idiosyncratic volatility, we were enabled to test whether high idiosyncratic volatility firms generated significantly higher CAARs than the low idiosyncratic volatility firms. The CAARs for firms with high pre-divest-ment idiosyncratic volatility were higher, however only significantly for the total and sell-off samples.
Thus, H1d was strongly accepted for the total and sell-off sample and weakly accepted for the spin-off sample.
The results using Tobin’s Q were similar. The CAARs for the low Tobin’s Q are significantly higher than the high Tobin’s Q for our total and sell-off sample resulting in a strong accept of H1e the samples. Surprisingly, the results for spin-offs showed that the firms with high Tobin’s Q have higher
CAARs than the group of low Tobin’s Q. However, the difference was insignificant, and thus H1e was weakly rejected for spin-offs.
H1f is related to the relative size of the divestment compared to the parent. Previous literature sug-gested that abnormal returns are affected by the relative size of the divested business unit. The larger the divestment, the larger abnormal return. The positive relationship between size and short-term stock return was highly significant for our total and sell-off sample, why H1f was strongly ac-cepted for these samples. Unexpectedly, the result for spin-offs showed an opposite correlation with no statistical significance, and thus H1f was strongly rejected for spin-offs.
At last, we tested whether the financial quality of a firm has influence on the short-term abnormal stock return. We used Altman Z-score to categorize our samples into low, medium, and high quality.
As reflected in H1g, firms of high financial quality were expected to generate higher short-term stock returns than those of low quality. In general, we found no connection between financial quality and short-term stock return of statistical significance across all samples, why H1g was weakly rejected.
Specifically, firms of low financial quality realized higher CAARs than firms of medium quality.
As described in the literature review, an increasing number of scholars have questioned the effi-ciency of capital markets suggesting that value creation should be measured over a longer period.
Therefore, we include two different long-term analyses measuring value creation on BHAR based on stock prices and annual ROA improvements based on different accounting measures.
Taking departure in the EMH and the equivocal results in existing empirical findings, H2 was formu-lated to test the long-term stock return expecting no returns significantly different from zero. How-ever, the analysis showed that firms engaged in both sell-offs and spin-offs realize significant positive BHARs in the three year holding period leading to a strong rejection of H2. The significant returns were found using two different benchmarks increasing the robustness of the findings. The results indicated higher long-term returns for firms engaged in spin-offs caused by the performance in both the parent and the subsidiary compared to firms divesting through sell-offs. Though, the difference was not found to be significant. Due to overlapping events for firms engaged in sell-offs, we are not able to conclude whether the abnormal returns are the result of one or multiple sell-offs. However, the findings indicate that firms continually divesting business units realize positive abnormal returns on the long-term.
Performance measures based on stock market returns are not always adequate measures of value creation. The existing literature suggested that corporate divestments might be motivated to mitigate operational inefficiencies enhancing operating performance, particularly relevant for overdiversified firms. To analyse whether the identified stock returns associated with corporate divestments are materialized in tangible improvements in operating performance, H3 was formulated expecting pos-itive changes in operating performance for both sell-offs and spin-offs. The findings showed that
firms engaged in sell-offs realize significant positive changes in operating performance measured on EBIT and EBITDA in the two years after completion. The results persisted when accounting for the industry development using a group of control firms. However, the positive change in return on cash flow from operations was insignificant. Based on the results from EBIT and EBITDA, we strongly accepted H3 for the total and sell-off sample. The results for spin-offs indicate a positive change in operating performance, though with less insignificance leading to a weak accept of H3.
Existing literature provide arguments for focus increasing divestments to realize larger and more significant improvements in operating performance compared to non-focus increasing divestments.
H3a was formulated based on the anticipation of focus increasing divestments are reducing the po-tential negative synergies of diversified firms. However, we found no empirical support for focus increasing divestments to realize larger improvements in operating performance, why H3a was weakly rejected across all samples.
In conclusion, we found overall strong indications of value creation related to European firms en-gaged in corporate divestments through spin-offs and sell-offs. However, we are not able to fully detect and capture the exact source of the value gains in corporate divestment. Though, based on our findings, corporate managers driven by shareholder maximising motives should continuously evaluate their portfolio of business units considering: “is our firm the most valuable owner for each of our subsidiaries?”. After such strategic review, managers can use spin-off to restructure ownership whereas sell-offs provide cash proceeds to invest in new projects or restructure financial leverage implying additional agency costs for investors.
9.1. Suggested further research
The analysis of corporate divestments completed by European, publicly listed firms in the past two decades enabled us to determine the shareholder value creation and the most relevant motives affecting the value creation. However, our thesis could not cover every aspect regarding corporate divestments implying several starting points for future research to be built upon.
The stringent sample selection processes implied that several divestitures were disregarded. As a result, our finding might not be generalisable for divestments due to the limited representation. De-spite of the precautions taken; it would be interesting to investigate whether the results are valid on a data sample constructed on more relaxed criteria. Specifically, the sample size of spin-offs in this thesis is small which affect the robustness of inferences. However, it could be interesting to expand the sample geographically, which would both increase the number of spin-offs and make an oppor-tunity to create new subsamples, e.g., Europe versus USA.
Future research could also investigate potential differences in value creation for different industries or effects of investor sentiments for hot industries affecting the short-term announcement effect. As illustrated in Section 7.3, firms operating within manufacturing are most present in the sample of our thesis. However, a focused analysis of divestments within a single industry might better capture specific value drivers of divestments for that specific industry.
The research design and methodology applied in this thesis was designed to answer the search question, why we have focused on determining the realized shareholder value creation of the in-cluded divestments. However, the identified value creation in corporate divestments might be af-fected by how the divestiture is initiated and structured. In real business practice, a firm may consider different divestiture types for the same unit at the same time, and then choose the best option. The effect of dual track processes is not included, but we find it to be interesting for future research to investigate how the shareholder value creation is affected by the orchestrated divestiture process.
The existing literature on shareholder value creation tends to investigate divestitures as isolated and unrelated corporate events. Oppositely, Brauer and Schimmer (2010) argues that divestitures should be analysed as strategically interelated events. Specifically, “selling a business is rarely a one-off activity” (Mankins, et al., 2008, p. 99) To account for value creation in firms engaged in multiple divestments, this thesis allows for overlapping events which means that transactions of the same firm are included if the transactions are not announced in the same [-10,10] event window. However, we were not able to identify which divestitures were strategically interrelated in a larger portfolio restructuring. Therefore, our thesis is grounded in a transaction-based perspective where the data sample has been gathered by identifying isolated transactions according to specified criteria. Another interesting approach for is to take a firm-based perspective where the data sample process is initi-ated by identifying a sample of firms and investigate their divestment activities. The firm-based per-spective might be more optimal in analysing potential value creation emerging from recurring divest-ment activities. A firm-based perspective would also allow for a comparison with other types of port-folio restructuring.