Branding a Merger: Implications for Merger Valuation and ...staff.washington.edu/epavlov/images/Branding a merger.pdfname and symbol), business-as-usual (both firms continue to operate

1

Branding a Merger: Implications for Merger Valuation and Future Performance

Isaac Dinner UNC

[email protected]

Jonathan Knowles Type2Consulting

[email protected]

Natalie Mizik∗ UW Foster School of Business

[email protected]

Eugene Pavlov UW Foster School of Business

[email protected]

January 18, 2019

∗ Corresponding author

Electronic copy available at: https://ssrn.com/abstract=1756368

2

Branding a Merger: Implications for Merger Valuation and Future Performance

The choice of post-merger corporate branding is a major strategic decision. It serves as a signal about the positioning and strategic intent of the new merged entity to the key stakeholders—customers, employees, and investors—affecting their ensuing behavior. We investigate the financial markets reactions and future operating performance implications of this decision. We classify merger transactions into three groupings according to the post-merger corporate branding: assimilation (the identity of the target company is discarded and it is rebranded with the acquirer’s name and symbol), business-as-usual (both firms continue to operate under their own corporate names and symbols), and fusion (elements of both corporate brands are maintained in a new brand). We find significant differences in the merger valuation across branding strategies. The stock market reaction to fusion is more positive than to assimilation and business-as-usual-branded merger announcements. Our analyses of post-merger sales, operating costs, and survival rates help explain these differences in merger valuation. Our analyses address selection and endogeneity issues, allowing for a causal interpretation of these findings. They provide support for the “brand-is-an-asset” versus the “signaling” perspective on the role of brands. Interestingly, we also find that the well-documented M&A mis-pricing phenomenon is primarily driven by the business-as-usual-branded mergers. Key Words: Corporate Branding, Mergers, Event Study, Calendar-Time Portfolio Analysis, Selection Bias, Treatment Effects Models, Endogenous Treatment Effects


3

Brands are valuable intangible assets. When managed effectively, brands can play a significant

role in creating value for organizations and their stakeholders. Academic research on branding has

provided insights into key issues such as the measurement of brand equity (Ailawadi, Lehmann,

and Neslin 2003; Fischer 2007; Goldfarb, Lu, and Moorthy 2009; Srinivasan, Park, and Chang

2005), brand portfolio management (Erdem and Sun 2002; Rao et al. 2004), brand naming

(Lowrey and Shrum 2007; Melnyk et al. 2012; Peterson and Ross 1972; Sood and Keller 2012),

brand extensions and co-branding (Bottomley and Holden 2001; Cao and Sorescu 2013; Lane and

Jacobson 1995), and brand valuation (Barth et al. 1998; Mizik and Jacobson 2009).

Most of the academic research on branding has focused on brand management under the

conditions of organizational steady state—that is, situations in which no disruptive changes are

occurring to company management, strategy, or business ownership (Bahadir et al. 2008). The last

few decades, however, have been characterized by consolidation and high levels of merger and

acquisition (business combinations) activity in many industries. Mergers remain popular in up and

down markets. According to Ahmed et al. (2018), 2018 is expected to be the third-biggest year on

record for merger and acquisition (M&A) deals, following the record of 2015 and the pre-

recession peak of M&A transactions in 2007.

Mergers are central to corporate growth, but they are also disruptive events that cause

company stakeholders—customers, employees, and investors—to reassess their relationships with

the new merged entity. Most academic research on mergers has focused on the role of internal

resources of the merging organizations (King et al. 2004). In contrast, we focus on a very public

dimension of mergers—the choice of the post-M&A corporate branding—which thus far has not

been examined. We argue that the choice of corporate branding is an important strategic decision

in business combinations because it communicates the positioning and strategic intent of the

merged entity and can influence the behavior of customers, employees, and investors. The loyalty

of these three constituencies is critical to the financial success of a merger.


4

Using a unique sample of large mergers undertaken between 1997–2006, we classify

merger transactions into three groupings according to the post-merger corporate branding strategy

selected: assimilation (target company is rebranded with the acquiring firm’s name and symbol),

business-as-usual (both firms continue to operate in the market under their own names and

symbols), and fusion (elements of both corporate brands are maintained in the new brand). We

assess the impact of this choice on firm value and future operating performance in the ten years

following merger completion (covering the 1998-2017 period) using inverse-probability weighted

regression adjustment (IPWRA) models. IPWRA models are ideal for these analyses as they

directly account for selection issues in our data (i.e., non-random choice of post-M&A branding)

and allow us to test and control for potential endogeneity and the violation of missing-at-random

assumptions. We undertake and report extensive sensitivity analyses of the methods and the data

sample to confirm the validity and temporal stability of our findings.

We find significant differences in the immediate market reaction to merger announcements

across our branding groups. Notably, firms undertaking assimilation and business-as-usual

branding experience significant negative market reaction at the time of the merger announcement,

whereas firms choosing fusion branding do not. Our analyses of post-merger operating

performance shed light on the underlying mechanism driving these findings: Overall, the

immediate and the long-term performance implications of fusion are significantly more positive

than those of assimilation and business-as-usual branding. Specifically, we find that the survival

and the future sales growth rates are higher and the future operating costs are lower for fusion than

for assimilation and business-as-usual-branded firms. Our results demonstrate a direct impact of

branding on operating performance measures and provide support for the “brand-is-an-asset”

argument over the alternative “signaling” explanation.

Interestingly, we also find a future negative valuation adjustment (drift) only for business-

as-usual-branded mergers, suggesting that the M&A mis-pricing phenomenon extensively

documented in prior research is primarily driven by business-as-usual-branded mergers. Portfolios


5

containing business-as-usual-branded firms underperform the market in the years following

merger completion, whereas portfolios of fusion and assimilation-branded mergers do not.

To the best of our knowledge, our study is the first to systematically examine the financial

value (i.e., stock market value) implications of corporate branding strategies in M&A and show

their impact on the future operating performance. In practice, branding decisions are rarely given

the careful consideration afforded to other aspects of merger transactions (Ettenson and Knowles

2006). We argue that this situation should change: Post-M&A corporate branding is an important

factor impacting the financial success of a merger.

Business Combinations

Firms engage in mergers to grow, diversify, gain access to new markets and resources, integrate

vertically, acquire R&D and patents, avoid direct competition, and reduce overcapacity. The

economics literature emphasizes efficiency-related motives for mergers, including economies of

scale and scope, leveraging various synergies across merging entities, attempts to create market

power by forming monopolies or oligopolies, and self-serving attempts by management to expand

its span of responsibilities and associated benefits (Andrade et al. 2001; Jensen 1993).

Empirical Evidence on Post-M&A Performance

Early M&A research focused on studying the financial consequences of mergers employed short-

window event studies to examine stock market reaction to merger announcements. These studies

generated two main findings: (1) shareholders of the acquiring firm earn, on average, zero or,

small negative abnormal returns; and (2) target company shareholders benefit from mergers and

accrue wealth gains at the time of the announcement. Mulherin and Boone (2000), for example,

report a small (but insignificant) negative abnormal return for acquiring shareholders and a 20%

positive abnormal return for the shareholders of the target firm in the three-day window around the

merger announcement date.

Subsequently, researchers began examining the longer-term financial consequences of

mergers using long-horizon event studies. This research documents an M&A mis-pricing

phenomenon—i.e., a long-term negative post-merger stock price adjustment (drift). Over time, this


6

negative stock price drift overwhelms any positive initial gains accruing to the target firms,

resulting in a net negative overall value effect of mergers. For example, Mitchell and Stafford

(2000) and Andrade et al. (2001) show that at the date of the merger announcement, the acquirer

experiences around -0.7% abnormal return and the target firm around 16% positive return. In the

subsequent three years, however, the equally weighted average abnormal returns to the

shareholders of the merged entity are -5.0%. Agrawal et al. (1992) report a similar finding of

negative long-term returns: The return to the merged entity is -2% after one year, -7% after three

years, and -10% after five years following merger completion.

Based on a review of 93 studies across the economics, finance, and management literature,

King et al. (2004, p.195) conclude that average “acquisitions either have no significant effect or a

modest negative effect on an acquiring firm’s financial performance in the post-announcement

period.” These findings raise important questions: If, in the aggregate, merger transactions fail to

create wealth, why do they remain so popular and what distinguishes successful mergers from

failures?

Explaining Post-Merger Performance

Multiple explanations have been advanced to explain the overall negative performance of mergers.

Research in economics and finance has focused on the misalignment of managers’ and

shareholders’ interests (agency theory: Jensen 1986; Kroll et al. 1997), management

overconfidence (hubris theory: Roll 1986), and the impact of transaction-specific variables (e.g.,

how it is financed, friendly vs. hostile, vertical vs. horizontal merger, King et al. 2004).

Management literature has focused on the market relatedness of the merging companies as a

determinant of post-merger performance (Andrade et al., 2001), resource complementarily of the

two firms (Harrison et al. 2001), and absorptive capacity (Zahra and George 2002).

Reviewing these alternative theories and explanations and empirical evidence, King et al.

(2004, p. 197-198) conclude that “despite decades of research, what impacts the financial

performance of firms engaging in M&A remains largely unexplained,” and suggest that

“researchers simply may not be looking at the ‘right’ set of variables as predictors of post-


7

acquisition performance.” Indeed, most of the M&A research has focused on the tangible asset

components of mergers and synergies of internal resources, manufacturing, distribution, and

product portfolios. Meanwhile, the growing importance of intangible assets is clearly evident as

the transaction prices relative to the book value of acquired companies continue to increase. To

cite one example, P&G paid $58.6 billion for Gillette in June 2005. This acquisition price

represented more than a ten-times multiple of Gillette’s less than $5 billion of tangible book value

($0.9 billion of net working capital and $3.6 billion of net property, plant, and equipment) and was

largely attributed to the strength of the Gillette brand name.

Marketing Assets in Mergers

Despite the widespread acknowledgment of the importance of marketing-related factors for

merger success (Becker and Flamer 1997), surprisingly little has been written about the marketing

factors in mergers (Homburg and Bucerius 2005). Marketing literature has almost exclusively

looked at mergers from a resource-based perspective, focusing on internal resources and the levels

and effectiveness of marketing capabilities: the extent and speed of marketing integration

(Homburg and Bucerius 2005), strategic emphasis (Swaminathan et al. 2008, Swaminathan et al.

2014), product vs. brand focus in acquisition announcement (Newmeyer et al. 2015), and product

capital (Sorescu et al. 2007).

One notable exception is Bommaraju et al. (2018) who have examined the impact of

M&As on sales force performance. This research shows that a merger with a poorer-image firm

dilutes salespeople’s organizational identification and impairs their performance. However,

questions related to how mergers affect market-based assets such as brand equity, customer

satisfaction, and customer retention are largely not addressed. This lack of research on external

factors and the stakeholders in mergers is surprising because mergers can change customer

attitudes and perceptions of a firm and its products and increase customer and employee defection

rates (Bekier and Shelton 2002, Thorbjornsen and Dahlen 2011). We seek to contribute to the

literature by focusing on the role of corporate branding in mergers and its performance

implications.


8

Corporate Branding and Mergers

Mergers are disruptive and transformative corporate events. They inevitably involve

organizational restructuring, changes to the management team and product offerings. Mergers

typically entail a rationalization of the combined portfolio of products and services,

reconfiguration of distribution arrangements, and attrition of employees and customers. The

uncertainty and changes a merger creates have major implications for the customers, employees,

and investors of the merging companies.

The choice of the corporate brand for the new merged entity is important because corporate

branding informs about managerial mindset and future behavior: It is a reflection of the internal

strategy for integration and re-structuring and a signal of management commitment to successful

integration. It can reduce uncertainty and help customers, employees, and investors make

inferences and form better expectations, potentially mitigating some of the disruptive effects of a

merger.

Branding Options in Mergers

At the time of the merger, managers of the merging firms have several alternatives for branding

the merged entity. They can choose to use the elements of either or both of the merging

companies’ identities (name and symbol), or they can create an entirely new identity. We

distinguish three main corporate brand strategy types, which we label as follows:

Assimilation – the identity of the target company is discarded entirely and all its operations

and products are rebranded with the name and symbol of the acquiring firm;

Business-as-Usual – the corporate brand identities of both companies are maintained and

they continue to operate under their own names and symbols in the product market; and

Fusion – elements from both merging brands are combined to form a new brand.

Each of these branding strategies communicates a fundamentally different message about

the merger and offers different perceived benefits, costs, and risks to customers, employees, and

investors.


9

1. Assimilation Branding

Assimilation branding has the advantage of simplicity and expediency. When one of the firms has

a stronger reputation than the other, customers and employees of the less reputable firm may view

assimilation branding as an upgrade. Investors may appreciate the clarity this corporate branding

provides about who is in charge and what the new entity stands for. Assimilation branding

communicates the benefits of the scale and presence that can be achieved through the adoption of

a single, well-known, unified identity across the range of the merged operations. It is an effective

strategy for consolidating market power and in circumstances where opportunities exist for cross-

selling and bundling products and services. Assimilation sends a message of increased strength

and power to the competitors and business partners, encouraging them to shift to a more

cooperative mode.

The downside of this strategy is that it fully discards all brand equity and associated

goodwill of the customers and employees of the acquired firm. It communicates the dominance of

one entity over the other and creates a clear sense that there is a winner and a loser in the merger

transaction. Unless handled sensitively and proactively, this corporate branding strategy carries

higher risks of disenfranchising the customers and employees of the acquired company who may

feel that their past history and relationship with the firm is being disregarded or erased.

Indeed, corporate brands are important not only to customers but also to employees and

firm management. Tavassoli et al. (2014) demonstrate the significance of the employee-based

brand equity in the executive pay context and Zamudio and Swaminathan (2016) further develop

this construct. Corporate brands communicate the identity, vision, and ideas, promote corporate

alignment around common values and goals, help establish the norms of the corporate citizenship,

energize and engage the employees. Corporate brands promote emotional and intellectual

engagement at work (Buil et al. 2016). Assimilation mergers tend to negatively affect employee

morale: Acquisitions double the turnover of employees and senior management teams (Hambrick

and Cannella 1993; Krug and Shill 2008) and stifle employment, wages (Li 2013, Siegel and

Simons 2010), and individual productivity (Bommaraju et al. 2018).


10

Under assimilation branding, both customers and employees of the acquired firm may feel

they are now in a new relationship that is not of their conscious choosing and thus are more likely

to use the opportunity to explore alternatives. Assimilation branding comes with the greater risk of

diminished future sales prospects and potential for higher operating costs because customers and

employees of the acquired company are more likely to defect.

2. Business-as-Usual Branding

Business-as-usual is a sound strategic choice when the merger is predicated on strategic or

operational benefits rather than explicit synergies in the customer and employee bases of the

merging firms. Under business-as-usual mergers, customers are unaffected and may be completely

oblivious to the ownership change as the corporate brand of the acquired company survives as a

divisional brand of the acquirer (e.g., Gillette becoming a part of P&G). Employees in business-

as-usual mergers are also typically less affected than those in assimilation mergers. The target firm

disappears for investors, but its brand lives on for consumers and employees.

Business-as-usual transactions are a good strategy for diversification and reducing

competitive pressures in the market, expanding the portfolio of products. The existing brand

equity of the target company is maintained, but no significant leverage of the customer base or

operating efficiency is sought after in a business-as-usual merger. Business-as-usual branding

sends a strong message about continuity and suggests the merger is a portfolio transaction. That is,

the value creation opportunities lie in reduced competition rather than in migration of customers

and employees to a relationship with a new brand.

One key downside of the business-as-usual branding is its high ongoing operating and

marketing cost for maintaining two distinct brands. The two customer bases of the merging firms

remain segregated (segmented) and are served by separate entities, thus, allowing to maintain the

sales growth, but at a higher cost. The other key downside is that business-as-usual branding does

not convey or promote unity and may be less conducive for the post-merger integration, impeding

the flow of potential benefits from operational or supply chain integration. As such, resistance to

integration and the operating costs might be rather high.


11

3. Fusion Branding

Fusion branding creates something new by explicitly combining the equities from the merging

companies, such as their corporate names (e.g., Thomson Reuters) or their symbols (e.g., United

and Continental Airlines merger, Table 1B). This branding communicates continuity and fusion of

the two entities and presents the merger as a transformative event for both firms.

The use of fusion branding communicates to all stakeholders that the merger is about the

combination of the capabilities and cultures of the two companies. Fusion branding requires more

careful planning and research and usually involves greater consideration and deliberation than the

other strategies. The advantage of fusion is that it sends a unique signal to customers, employees,

and investors that senior management teams of both firms have actively considered their interests

and have made a commitment to working together to make the merger a success.

A disadvantage of fusion branding is that it is initially more costly to implement. Under

fusion branding, all operations of the merging entities need to be rebranded as compared to the

partial rebranding required under assimilation (i.e., only one entity is rebranded) and very limited

rebranding (if any) required under business-as-usual.

Table 1A summarizes key differences across branding strategies and Table 1B presents

examples of these branding strategies from the airline industry and examples of variations in

fusion branding. In summary, assimilation branding is the most taxing strategy from the customer

and employee perspective, but it makes a strong and clear statement of strategic and operational

unity of the new merged entity. Business-as-usual is the least disruptive strategy for customers and

employees. It has low immediate rebranding costs but is the most costly in the long run (as both

corporate brands are maintained) and the least conducive for post-merger integration of the

merged firms. The fusion strategy, while inclusive and potentially appeasing to customers and

employees of both firms, has the highest initial marketing costs, as operations of both merging

entities are rebranded following the merger.


12

Hypotheses

Branding may contribute directly to the merger success because it is a value-generating market-

based asset impacting customer behavior (i.e., it has demand-shifting effect) and employee

behavior (costs-shifting effect). We refer to this mechanism as “brand-as-asset” perspective.

Effective branding can help the merged entity enhance and/or preserve brand equity for customers

and employees of the merging firms, expand appeal of its products to new segments, generate new

incremental value to customers and employees through improved image and better marketing of

firm products, and streamline the integration and acculturation process. As a result, the choice of

corporate branding in mergers can affect the stock market valuation (i.e., the equilibrium

expectation of future performance) of the merged firm. Under the brand-as-asset view, differences

in valuation across branding strategies are a direct reflection and consequence of the expected new

value creation (increased sales and/or decreased costs).

Signaling is another mechanism that can potentially explain differences in market

valuation across branding strategies. The choice of the branding strategy in mergers can serve as a

signal to the investment community of how well the involved companies have thought through the

merger transaction and post-merger integration. It also signals the strategic direction and vision,

managerial intent, and commitment to successful integration of the merging organizations into a

single new entity, thus affecting the perceived risk of the merger and the discount factor investors

use for valuing the expected future cash flows. The signaling and brand-as-asset effects are not

mutually exclusive and might be present simultaneously. We discuss their implications to form

our hypotheses.

Stock Market Valuation of the Post-M&A Branding

The assimilation branding strategy discards all brand equity and the embedded customers’ and

employees’ goodwill of the target firm. As such, it can depress sales and initially accelerate

operating costs. Business-as-usual branding preserves the brand equity of both firms but does not

seek to enhance it, does not encourage integration, and limits potential synergies. Sales growth

would be unaffected, but integration is impeded and operating costs might rise in the long run.


13

Fusion branding seeks to preserve, enhance, and leverage brand equity of both merging firms and

to facilitate their integration. It is the most expensive choice initially, but sales might accelerate

and operating costs decline in the future. Thus, from the brand-as-asset perspective, fusion

branding has potentially the most positive expected future cash flow implications (increased sales

and decreased long-term costs). From the signaling perspective, fusion, again, is the strongest and

most credible (as it is the most expensive) signal of managerial commitment, followed by

assimilation (clear signal of unity), and then business-as-usual branding (with no clear signal of

commitment to integration, higher uncertainty leads to higher discount rates for valuing expected

cash flows, resulting in lower values). As such, both the signaling and the brand-as-asset

perspective favor fusion branding:

Hypothesis 1: The immediate market reaction to merger announcements is more positive for acquirer firms pursuing fusion corporate branding than for the other firms.

Under the efficient market hypothesis the change in the market valuation at the time of the

announcement reflects the changes in the expectations of the future performance (it is an unbiased

expectation of the incremental future discounted cash flows to the firm). Testing H1 and

estimating the market reaction to the choice of post-M&A branding provides a summary estimate

of the expected future performance impact, but does not clearly distinguish whether the market

reaction is due to differences in the expected future operating performance (brand-as-asset view)

or due to differences in perceived risk (signaling view) across branding strategies. With the long

time-series data of post-M&A operating performance and stock returns we can assess directly the

role of these two potential mechanisms. Finding a differential future operating performance across

branding strategies would provide direct support for the brand-as-asset argument. Finding

differences in risk profile would support the signaling argument.

Post-M&A Operating Performance

Under the brand-as-asset view, our three branding strategies have different implications for the

firms’ key stakeholders and these differences will be reflected in the differential operating

performance. Fusion branding seeks to preserve and enhance customer-based brand equity. As


14

such, in the long run, we would expect enhanced sales for firms selecting fusion branding.

Business-as-usual also preserves brand equity for the customers and sales would continue to grow

unimpeded. Assimilation, on the other hand, discards the brand equity of the acquired firm and

runs a high risk of alienating its customers. Sales performance of assimilation branded mergers is

likely to suffer, particularly, in the immediate period following merger completion:

Hypothesis 2: Merging firms adopting fusion and business-as-usual branding have higher sales growth in the post-merger period than assimilation firms.

The three branding strategies also have differential implications for the employees and the

operational integration of the merging firms. These differences will be reflected in the differential

operating costs in the post-merger period. Fusion seeks to preserve and enhance employee-based

brand equity and to facilitate the integration of the two merging firms. We would expect low

future operating costs for firms selecting fusion branding. Assimilation disenfranchises the

employees of the acquired firm and, as such, increases costs in the near term, but it facilitates

integration and alignment of the merged firm operations, reducing the costs in the long-run.

Business-as-usual preserves employee equity but it does not encourage integration into a single

entity and does not allow the merged firm to capitalize on potential synergies. A failure to

successfully integrate and streamline operations and management control after a merger may to

come with the lack of clear strategic direction and greater dysfunction at the top of the

organization and can result in higher operating costs for the merged entities in the long-run. As

such, we would expect:

Hypothesis 3: Merging firms adopting assimilation branding initially exhibit the highest operating cost increases, but in the long-term, the business-as-usual-branded firms have the highest operating costs growth. Risk Profile

Past research found positive stock market reaction to corporate name changes (e.g., Horsky and

Swyngedouw 1987) and has argued that new name per se does not increase demand for the firm

products (i.e., arguing against the “brand-is-an-asset” perspective). Rather, the name change serves

as a signal of managerial commitment and a sign that the company is undertaking other


15

organizational changes aimed at improving firm performance thoughtfully and deliberately. Under

the signaling view, both fusion and assimilation branding send a strong signal of managerial

commitment to post-M&A integration. Business-as-usual branding, on the other hand, sends of

signal of independence of the merging firms and no commitment to integrating operations. While

the product line diversification inherent in the business-as-usual strategy can reduce risk in

product-market, greater uncertainty about the leadership and future strategy of the merged firm

may lead to higher discount rates and higher risk of the business-as-usual-branded mergers:

Hypothesis 4: Merging firms adopting business-as-usual branding have a riskier profile than the other firms. Stock Market Mis-valuation of the Post-M&A Branding

Under the efficient market hypothesis, the expected differences in the post-M&A operating

performance or risk across the branding strategies (if any exist) drive the differential market

reaction to M&A announcements. If the market correctly anticipates future financial consequences

of a merger, then the initial market reaction will fully reflect the unbiased expectation of the

change in its discounted future cash flows. Under the assumptions of market efficiency and full

information, no future-term systematic adjustment (drift) in the valuation of the merged entity

should occur.

Past studies, however, have consistently documented a significant downward adjustment in

the valuation of firms undertaking a merger (e.g., -7% in 3 years post-merger completion, Agrawal

et al. 1992). Researchers have interpreted this evidence to suggest that the market might not be

able to properly price mergers at the time they occur and is, on average, overly optimistic about

the future prospects of mergers. This initial optimism is corrected over time with systematic

downward adjustment in the valuation of the merged firms. Past research also suggests the market

has difficulty in pricing certain types of strategic decisions and intangible assets (e.g., Daniel and

Titman 2003; Eberhart et al. 2004). A corporate brand might be one of such difficult-to-price

assets. Research has not considered whether the initial optimism and overvaluation of mergers and


16

the subsequent downward correction are associated with the choice of corporate branding strategy.

We investigate this question here.

We do not propose formal hypotheses on the direction of the drift because no theory

predicts why a particular type of branding might be over- or undervalued initially. If the market

does not properly anticipate the implications of branding strategy in mergers, a future-term

adjustment in the market valuation would occur to correct the initial mis-valuation and it may

differ across post-M&A branding strategies. That is, rather than reacting fully at the time of the

merger announcement, the market might continue to systematically adjust the valuation of the

merged enterprise in the future after observing customer and employee reaction and the resulting

cash flows. We undertake a drift study to assess whether the market initially under or over-values

branding choices.

Data

We combine information from multiple data sources to compile our research data set. We use the

SDC Platinum database for information on merger transactions and their characteristics. We

obtain daily stock returns data from CRSP and accounting data from Compustat database for

1997-2017. Fama-French and Carhart daily risk factors come from Kenneth French’s web data

library. The data for our post-M&A branding classifications come from Type 2 Consulting (T2).

T2 used a three-step procedure to create a unique data set on corporate branding for merger

deals announced during 1997-2006. Following the initial announcement of a merger, T2 analysts

first searched secondary data sources (company websites, press releases, media reports) to

establish the pre-merger corporate branding information for both companies (names and symbols).

Then, they sent surveys to the merging companies and followed up with a telephone interview

with those who responded to the survey. T2 data has excellent coverage for large merger

transactions. With smaller-size transactions pre-merger branding information (particularly for

brand symbols) was more challenging to acquire and the response rate to survey and interview

requests were notably lower. T2 pursued and compiled data for smaller-size mergers until the

response rates dropped significantly and the search and follow-up costs became prohibitive.


17

Merging these five data sources generated a set of 210 merger transactions that fit our

criteria for inclusion (i.e., both merging entities are US-based public companies listed in CRSP

and Compustat). Table 2A presents the distribution of our mergers data across industrial sectors.

98% of the mergers in our dataset were “friendly” transactions (only three mergers in our dataset

are tagged as “friendly to hostile” or “hostile” in the Capital IQ database).

Table 2B reports our sample coverage by the deal transaction value. Comparing our

research sample to all transactions in SDC Platinum involving U.S.-based firms in that time frame,

we find that our sample accounts for nearly all transactions over $25 billion and for over 80% of

transactions over $10 billion. From a transaction value perspective, our dataset accounts for 55%

of all transactions over $100 million.

Descriptive statistics for the target and acquirer firms are presented in Table 2C (the target

and acquirer designations come from SDC Platinum). These data come from the last annual report

filed by the merging firms prior to the merger. Of the 210 mergers in our sample, 118 chose

assimilation, 49 business-as-usual, and 43 fusion branding. In only three cases, the merged entity

created a completely new brand identity. We included these three cases into the fusion branding

group, but as we later report in our sensitivity analyses section, our results are not sensitive to this

choice. Table 2D presents descriptive statistics for our key research variables by branding strategy

and their definitions.

As Table 2D shows, there are significant differences across our branding groups on some

dimensions, suggesting a non-random assignment into the strategy groupings and calling for

explicit modeling potential selection issues. For example, we note a significant difference in the

proportion of firms choosing assimilation branding in horizontal mergers (i.e., involving two firms

within the same industrial grouping). Another notable difference across the three groupings is in

their propensity to explicitly acknowledge their intended future branding strategy in the text of the

official merger announcement. The intended post-M&A branding is typically discussed in the

media and other public data sources immediately following the announcement (i.e., it is public

information at the time of the merger announcement). However, only 11% of firms undertaking


18

assimilation branding and 16% undertaking business-as-usual branding explicitly referenced their

corporate branding decision in the merger announcement. In contrast, 35% of firms choosing

fusion branding discuss it in the merger announcement itself.

Empirical Analyses

Testing Hypothesis 1. Immediate market reaction to merger announcement

Traditional event study analysis

Under H1, we would observe a more positive market reaction to merger announcements for firms

pursuing a fusion branding strategy. We can test H1 using event study methods (MacKinlay

1997). A standard event study analysis proceeds as follows. First, we compute the abnormal stock

return (ARit) for firm i, day t as

ARit = Retit – E[Retit], where (1)

Retit is the raw return for firm i on day t and E[Retit] is the expected return. We use the pre-

announcement period beginning 12 months (252 trading days) before and ending one month (21

trading days) before the merger announcement date and the four-factor asset pricing model (Fama

and French 1993, Carhart 1997) to compute expected returns. That is, we first estimate the

following model for each firm i in the [-252; -21] window preceding a merger announcement q:

Retit-RiskFreet = αqi + βmkt,qi(RetMktt–RiskFreet) + βSMB,qiSMBt

+ βHML,qiHMLt + βUMD,qiUMDt + εit, where (2)

RiskFreet is the risk-free rate, RetMktt is the market return, SMBt is the difference in returns

between small and large firms, HMLt is the difference in returns between high- and low-value

firms, and UMDt is the Carhart (1997) momentum factor.

Next, we use the estimates of market ( β̂ mkt,qi), SMB ( β̂ SMB,qi), HML ( β̂ HML,qi), and UMD

( β̂ UMD,qi) risk factor loadings to compute abnormal returns ( itAR ) for each firm i and day t around

the merger announcement q, as the difference between the actual and expected return. Table 3A

reports ARs for the acquirer firms for ten days before and ten days after the merger announcement.

We observe no significant advanced leakage of the news for any of the groupings or for our

sample as a whole. Consistent with past research, we find small negative but statistically


19

significant market reaction to the merger announcement clustered tightly around the

announcement date. We examine the ARs to identify appropriate event window and aggregate ARit

over the duration of the event window to compute cumulative abnormal returns (CAR) for each

firm i and event window [t1; t2] as ∑=

=2

1

.),( 21

t

tiiq ARttCAR

tt

Table 3B reports CARs in the [-1; 0] window. The acquirer firms realize about 2% negative

return around the announcement date. We report two tests of significance for the average group

CARs (both are common in the literature): the standard t-test and Corrado (1989) non-parametric

event study rank test statistic θ. The implied statistical significance of these returns is greater with

non-parametric Corrado than with standard t-test. Careful examination of the distributional

properties of abnormal returns in our data reveals normality of the mean and skewness, but higher

than normal kurtoses in the data. Thus, t-stats underestimate the significance (are conservative).

Table 3B shows differences in the market reaction across the three brand strategy

groupings. The acquirer firms pursuing assimilation branding realize significant negative returns

of -2.61% and the business-as-usual-branded mergers also realize a significant but somewhat less

negative returns of -1.92%. We observe no negative market reaction to mergers pursuing a fusion

strategy. The cumulative abnormal returns to these firms are not significantly different from zero.

In line with past research (Mulherin and Boone 2000), we find significant positive average

cumulative abnormal returns of 20.62% accruing to the target firms in the [-1; 0] event window

and find that merger premium (the difference between the price offered for the target and its stock

price before the merger announcement) is the major driver of this positive market reaction to

merger announcement for the target firms (p<.001). But we find no significant differences (p=.23)

in market reaction across our three branding groups ([-1; 0] CARs to the target firms are 23.06%,

17.20%, and 17.91% for assimilation, business-as-usual, and fusion-branded mergers,

respectively).

Modeling selection issues and additional influences in the event study analysis


20

One concern with traditional event study analyses above pertains to potentially non-random choice

of branding strategy. Post M&A branding is likely a deliberate decision by merging firms. As

such, a selection bias might be affecting findings reported in Table 3B. Simply comparing mean

CARs across our groupings does not account for the possibility that certain types of M&As or

firms might be more likely to choose a particular branding strategy, and it might be the

characteristics of the merging firms, rather than the branding strategy, that are driving the

observed differences in market reaction. Further, some merger-specific factors correlated with the

branding strategy might be driving or masking significant findings. Failure to control for these

factors can result in incorrectly attributing their effects to branding strategy. We need to model

these effects to properly estimate the impact of branding strategy.

In the last two decades, much progress has been made in developing the econometric and

statistical tools for estimating causal effects (for a review of the causal inference and potential

outcomes framework and methods see Imbens and Rubin 2015). These methods are also gaining a

wider acceptance in marketing (Rossi 2018) and they have been adopted in M&A research (Li

2013, Li and Prabhala 2007). Imbens and Wooldridge (2009) review recent advancements and

provide practical recommendations for empirical researchers seeking to identify causal effects of a

“program” (in our case, post-M&A branding strategy). We follow their advice and use an inverse

probability weighted regression adjustment model (as we later show IPWRA modeling framework

is best fitted for our research problem and data).

IPWRA estimators model both the outcome (market reaction) and the “treatment” (brand

strategy choice) to account for the nonrandom treatment assignment. That is, we account for the

likelihood that merging firms choose a specific branding strategy, and condition the outcomes of

interest (CARs) on the probability that the firm has chosen a specific strategy. We choose the

IPWRA over alternative popular estimators, e.g., matching, for three main reasons. First, the large

M&A transactions we study are relatively rare and unique and our dataset covers almost the

entirety of the large transactions. As such, it is impossible to find an appropriate match. Second,

IPWRA tends to perform better in small samples than matching and other approaches (Busso et al.


21

2014). Third, IPWRA is consistent, asymptotically efficient, and double-robust (i.e., only the

propensity score model or the regression adjustment need to be correctly specified for the entire

system to be consistent, Wooldridge 2010).

The first step in IPWRA estimation involves modeling the propensity score (the likelihood

of choosing one of the three alternative branding strategies). We model this choice using a

multinomial logit model based on the acquiring and target firms’ characteristics, their respective

industries’ characteristics, and merger-specific characteristics:

Yqj=αj+ΣβM&AXM&A+ΣβAcqFirmXAcqFirm+ΣβTFirmXTFirm+ΣβAcqIndXAcqInd+ΣβTIndXTInd+ εqj, (3)

where j refers to one of the potential alternative (assimilation, fusion, business-as-usual), q refers

to a specific M&A case, and Yqj=1 if M&A case q chooses branding option j. XM&A is a set of

merger-specific characteristics, XAcqFirm is the set of the acquirer and XTFirm is the set of the target

firm characteristics, XAcqInd and XTInd are the acquirer and the target’s industry characteristics.

We follow Hirano and Imbens’ (2001) procedure for selecting which variables to include

in the propensity score model: We start with a large set of possible covariates and implement a

stepwise approach to ensure only relevant covariates are included. We begin estimation with all

factors listed in Table 3D, but find that only some are significant predictors of the post-M&A

branding strategy. The final model resulting from our stepwise estimation is presented in Table 4.

Horizontal mergers are more likely to use assimilation branding. Acquirers with house-of-brands

portfolios are much more likely to choose business-as-usual strategy. A mention of the new brand

in the merger announcement is more likely with fusion strategy. Acquirer and target sizes and the

target’s advertising intensity also help predict branding strategy choice. None of the other potential

explanatory factors we considered are statistically relevant in our sample.

The second step of IPWRA uses estimated probability weights of choosing a particular

branding strategy to weigh observations in the regression adjustment model explaining abnormal


22

returns. The process of weighting effectively adjusts the data to create a pseudo-population such

that there is no confounding on observables and the weighted averages reflect averages in the true

population, thus eliminating selection bias. We estimate the following model:

CARq=αq+λXq+ εq, where (4)

each merger case q is weighted by the inverse probability that the merger would choose a

particular strategy and Xi is a set of merger-specific characteristics.

Table 5 reports results of our regression adjustment model for CARs [-1;0]. Table 5 Panel

A presents the potential outcome means for each branding strategy (i.e., the mean potential

outcome that an individual case would obtain if it chose branding strategy j) and average treatment

effects (i.e., the effect of choosing option j versus an alternative j’). We see a pattern similar to that

reported in table 3B. Assimilation and business-as-usual-branded mergers realize significant

negative returns while fusion branded mergers do not. Addressing selection issues and including

the regression adjustment components (i.e., additional controls whose estimates are reported in

Table 5 Panel B) slightly changes the pattern of the estimated average group returns. We now

observe a somewhat larger negative reaction to business-as-usual branding. The differentials

between fusion and assimilation and between fusion and business-as-usual are both now more

significant. The choice of fusion strategy is associated with a 2.3% greater abnormal returns as

compared to the assimilation and business-as-usual-branded mergers combined (p=.01). As such,

we find full support for Hypothesis 1.

Table 5 Panel B reports the estimates for the covariates included in the regression

adjustment stage. Horizontal mergers using assimilation branding receive a more negative market

reaction. House of brands acquirers using business-as-usual branding receive a more positive

market reaction. Announcing the intended branding in the business-as-usual mergers has a

marginally negative impact on acquirer’s CARs. Merger premium has a positive effect on CARs

for assimilation-branded mergers. These results highlight the fact that while these factors are


23

relevant in explaining market reaction, they are incremental to the effects of branding strategy and

do not drive our findings of differential market reaction to post-M&A branding.

We undertake several tests to ensure the IPWRA model is appropriate for our data.

IPWRA models require several assumptions and the findings reported in Table 5 can only be

interpreted as causal if these assumptions hold. We test whether conditional independence

(selection on observables), overlap, and IID assumptions hold in our model 4 and find that they

do. We test for endogeneity (correlated unobservables in treatment and outcome equations) with

the control function method (Wooldridge 2015) and find no endogeneity. We examine the

estimated densities of choosing a particular type of branding and find that the densities are massed

over the same region suggesting the overlap assumption holds. We find no dependencies across

events, suggesting IID assumption holds. Finally, we undertake a covariance balance tests and, as

expected, find no significant differences across groups. All sensitivity tests (available on request)

suggest IPWRA models are appropriate for our data.

Testing Hypotheses 2 and 3: Post-M&A operating performance

H2 and H3 argue that a brand, if it is a value-generating asset, will significantly affect subsequent

M&A operating performance (i.e., resulting changes in customer and employee attitudes,

preference, loyalty, and engagement would affect future revenue and costs). Because a merger is a

major structural change to the entities we study, past values of the data series cannot be used to

form appropriate instruments and standard dynamic panel data methods (e.g., Arellano and Bond

1991) cannot be used to test post-M&A operating performance differences across our groups.

Therefore, we follow research in finance (e.g., Healy et al. 1992, Harford et al. 2012) and

benchmark the post-merger operating performance on the combined performance of the merging

firms in the final year before merger completion. We compute a measure of post-M&A sales

growth for firm i in post-M&A year t relative to the pre-M&A sales base of the two merging firms

as the difference between the sales of the merged firm i in year t and pre-M&A total sales of the

merging firms (target and acquirer) scaled by the pre-M&A total sales of the merging firms:

ΔSalesit =(Salesit-(Salestarget,i0 + Salesacquirer,i0)) / (Salestarget,i0 + Salesacquirer,i0). Similarly, we


24

compute the operating costs growth as ΔOpCostit =(OpCostit-(OpCosttarget,i0 +

OpCostacquirer,i0))/(OpCosttarget,i0+OpCostacquirer,i0).

Recent work in finance has extended this post-M&A performance measurement approach

and has utilized the potential outcomes framework and methods to study post-M&A performance

(e.g., Li 2013). We follow this approach. We use IPWRA modeling to control for the non-random

choice of post-M&A branding strategy and examine differences in operating performance (sales

and cost growth) over ten years following merger completion. The ten-year post-M&A testing

period potentially adds an additional methodological complication. IPWRA approach is consistent

only if the “missing-at-random” assumption holds. We may have an issue of non-random missing

data due to firm attrition from the sample at longer horizons. As such, we need to ensure that the

post-M&A survival rates do not differ across our groups.

Assessing the missing-at-random assumption

A key assumption of the basic IPWRA model is that data are missing at random. A situation

where observations are missing at random results in loss of efficiency, but not in the loss of

consistency. A non-random loss of data can lead to significant biases in the estimates. This

assumption might be violated in our data if the survival rates in the ten years following a merger

differ across our branding strategies.

Table 6 Panel A presents survival rates data for our three branding groups in the ten years

following merger completion. Firms undertaking business-as-usual branding have the lowest and

firms undertaking fusion branding have the highest survival rates. A basic test of proportions

indicates that the survival rate differential between fusion and business-as-usual is significant in

years eight, nine, and ten (p<.05). The survival rate differential between assimilation and business-

as-usual is also significant in years eight and ten (p<.05). As such, the missing-at-random

assumption might not hold in our data. Ignoring the differences in survival rates can lead to

overestimating performance for business-as-usual (lowest survival likelihood) and

underestimating performance for fusion-branded firms (highest likelihood of survival).


25

We use Cox (1972) proportional hazard model to formally assess whether the observed

differences in survival are associated with the choice of post-M&A branding or can be explained

by some other factors. The Cox (1972) proportional hazard model is popular both in medical

research (to model patients’ survival) and in marketing applications (e.g., Hitsch 2006, Hu and

Van den Bulte 2014). It allows obtaining consistent estimates in situations where data are right-

truncated. It is preferred over logistic models which ignore survival time and censoring

information. Right-censoring is important in our setting as we can observe our M&A firms up to

the end of our study period (June 2017) and more than half are still surviving at that point (i.e., are

listed as active in CRSP). The Cox proportional hazards model is specified as follows:

ℎ(𝑡𝑡|𝑋𝑋𝑖𝑖, 𝑆𝑆𝑖𝑖) = ℎ0(𝑡𝑡)exp (𝛽𝛽𝑆𝑆𝑖𝑖 + 𝛾𝛾𝑋𝑋𝑖𝑖), where (5)

the dependent measure is time (duration) before delisting (censoring status=0) or the end of

observation period (censoring status=1), measured as the number of months the firm is listed in

CRSP since the M&A completion date. ℎ0(𝑡𝑡) is the baseline hazard function, 𝑆𝑆𝑗𝑗 is an indicator of

the branding strategy, 𝑋𝑋𝑖𝑖 is a set of M&A, acquirer, and target-specific characteristics (acquirer

and target’s size, profitability, advertising intensity, HHI, and industry average revenue growth;

merger premium, house-of-brands strategy indicator, name announcement indicator, horizontal

merger indicator; ratio of the target’s to the acquirer’s assets, acquirer industry dummies, and time

dummies). We utilize a stepwise procedure for model selection and report final results in Table 6

Panel B.

Table 6 Panel B reports estimation with fusion-branded M&As serving as a base. We find

that business-as-usual-branded mergers have a significantly higher hazard (risk of death) and the

assimilation mergers have marginally higher hazard than the fusion-branded mergers.1 In

addition, we find that business-as-usual mergers also have a marginally higher hazard rate than

1 We undertake sensitivity analyses to confirm our findings. A key assumption of the Cox proportional hazards model is that hazard ratio is constant over time. We validate this assumption using test based on scaled Schoenfeld residuals (Grambsch and Therneau 1994) and cannot reject H0 that hazard ratio is constant over time (p=0.80). We also test alternative survival models with differing assumptions about the parametric forms of survival functions (Weibull, exponential, Gompertz, loglogistic) and find the same pattern of results.


26

assimilation-branded mergers (p<.1). Clearly, the missing-at-random condition is not satisfied.

This finding calls for explicitly modeling the attrition process in order to correct for attrition bias.

Controlling for non-random attrition in the IPWRA framework

Cheng and Trivedi (2015) suggest estimating an auxiliary “attrition” model which is a probit

model where indicator I{Y≠N/A} is regressed on relevant observables. The estimation of the

attrition model generates the “attrition hazard” term. Including the predicted attrition hazard as an

additional covariate in the outcome equation corrects estimates for the attrition bias. We follow

Cheng and Trivedi (2015) and estimate attrition hazard separately for each time horizon (1, 2, …

10 years post-M&A completion) of sales and operating cost changes in the ten-year post-M&A

period we study. That is, the attrition hazard is time-varying.

We assess differences in sales and cost growth rates over the ten years following merger

completion with IPWRA models. Similarly to testing H1, we use propensity score models

(reported in Table 4) to correct for non-random strategy selection as described previously. We

include the year-specific attrition hazard correction and identify a relevant set of control variables

to be included into the regression adjustment stage. We estimate our IPWRA models for each one

of the ten years following merger completion separately. Presence of attrition hazard in the

regression adjustment assures our estimates of interest are free of attrition bias.

Table 7A reports key results of this estimation—potential outcome means and average

treatment effects—for sales growth for each of the ten years following M&A.2 In the first three

years following a merger we see no significant differences in sales growth across branding

strategies. After year three, however, fusion significantly over-performs assimilation and over-

performs business-as-usual after year five. Contrary to our H2, we find no consistent and

significant differences between assimilation and business-as-usual sales growth rates.

Table 7B reports results of our estimation for operating expenses for ten years after a

merger. We observe declining operating costs for fusion and increasing operating costs for 2 In the interests of space and because these results are not central to our research question, we report full regression adjustment results only for year ten. Table 7 reports estimates for the regression adjustment covariates in the sales model and for operating costs model. Notably, the attrition hazard is significant in these models.


27

assimilation and business-as-usual. In the first two years after M&A we find that fusion-branded

mergers have a significantly lower operating cost growth compared to assimilation, but in the later

years this difference disappears. Fusion also has a significantly lower cost growth than business-

as-usual-branded mergers in year two and in the later period of study. The assimilation-branded

mergers too exhibit lower cost growth than business-as-usual mergers in the later years. As such,

we have full support for H3.

We undertake several sensitivity tests (similar to testing of H1) to ensure the IPWRA

modeling is appropriate for our data and its assumptions hold (all available on request). We find

no reason to question the validity of the approach. Importantly, we undertake tests for endogeneity

(correlated unobservables in treatment and outcome equations) for each model and each time

horizon with control function method (Wooldridge 2015) and find no endogeneity.

In sum, our analyses of operating performance provide evidence supporting the brand-as-

asset explanation: post-merger branding affects future sales and operating costs. These findings

align with and are fully consistent with the more positive stock market reaction to fusion versus

assimilation and business-as-usual branding reported in our tests of H1: The more positive future

revenue and cost prospects of fusion firms are reflected in their better valuations.

Testing Hypothesis 4: Risk Profile Differences

To test H4 we examined differences in risk profiles of our three portfolios. Past research in

marketing has studied financial risk implications (e.g., Bharadwaj et al. 2011, Han, Mittal, Zhang

2017, McAlister et al. 2007, Rego et al. 2009). Similarly, we examined three common measures of

risk—total, idiosyncratic, and systematic—derived from 3-months, 6-months, 1, 2, 3, and 5-year

horizons after M&A completion. Total risk is measured as the variance of portfolio return for the

respective time period. We calculate idiosyncratic risk as a standard deviation of the residuals

from a Carhart’s (1997) four-factor model estimated on daily data. Systematic risk is the estimated

market Beta (βmkt) from the same model.

We find no notable differences in the total, systematic, and idiosyncratic risk across all

horizons we examined (Table 8). We find no evidence to support H4 and the argument that the


28

differences in the market reaction to merger announcements can be attributed to differential risk

profiles of our branding groups. The more positive value of fusion branding does not stem from a

higher risk profile of fusion-branded firms. To the contrary, the total portfolio risk tends to be

lower (albeit insignificantly so) for fusion than for business-as-usual and assimilation portfolios.

The systematic and idiosyncratic risk profiles do not produce clear and consistent patterns across

our three portfolios either. As such, we find no support for the signaling perspective on the effects

of branding in M&As.

Delayed Market Reaction: Time-Varying Calendar-Time Portfolio Analysis

Figure 1 presents raw calendar portfolio returns (3-year hold) for our three branding portfolios. It

tracks the investment of $1,000 into each of the portfolios for the duration of our study period and

depicts the value of each portfolio at different points in time. The compositions of portfolios

change throughout the study period as new mergers are completed and we place equities into their

respective portfolios for a three-year period. At the end of the three-year post-merger period, we

remove the equity from the portfolio. On average, merged firms undertaking fusion branding

appear to realize more positive returns than firms undertaking assimilation and business-as-usual

branding through most of the observation period. Raw returns, however, are not appropriate for

testing the differences in portfolio returns as differences in the risk profiles of the portfolios we

designed might drive some of the differences we observe in Figure 1.

Figure 2 presents average buy-and-hold risk-adjusted (i.e., abnormal) stock returns for

firms in our sample. All mergers are aligned at time zero by the date of merger completion, and

their abnormal stock returns are tracked from three days after to three years after merger

completion. We observe immediate separation of average fusion merger returns as they track into

the positive return region. Three years after the merger, firms undertaking fusion branding realize

on average an 11% positive abnormal return. Assimilation and business-as-usual average

abnormal returns immediately edge into the negative. One year after the merger completion, the

business-as-usual mergers realize less negative returns than the assimilation mergers, but two and

three years after the merger they have greater negative returns than assimilation-branded mergers.


29

Three years after a merger, firms undertaking business-as-usual branding realize on average a -

26% return, and firms undertaking assimilation branding realize a -19% abnormal return. This

pattern suggests business-as-usual mergers might underperform assimilation mergers in the long

run, but this underperformance is not apparent until several years after the merger completion.

Although insightful as an illustration, the buy-and-hold abnormal returns presented in

Figure 2 do not directly control for potential cross-sectional dependency and clustering of merger

events. Cross-sectional dependency, if present and not explicitly modeled, might lead to erroneous

inferences about the significance of estimated effects. We undertake a formal test for the presence

of a drift and assess differences across the groups using the calendar-time portfolio approach

(CTP). The CTP approach is the traditional method for assessing delayed market reaction. It is

advocated by Fama (1998) and is “robust to the most serious statistical problems” (p. 291,

Mitchell and Stafford 2000). It has been used in marketing (e.g., Sorescu et al. 2007, Gielens et al.

2008). This method is particularly appropriate for empirical situations in which events are

clustered in time and cross-sectional dependency might be present. This might be the case in our

data as mergers tend to come in waves.

The calendar-time portfolio approach involves creating a portfolio of securities based on

some attribute of interest, estimating a risk model for the portfolio, and testing for the significance

of the intercept in the estimated risk model. Securities are placed into the portfolio after the

information about the attribute of interest becomes public and the market has had sufficient time to

react to this new information (typically just a few days). Securities are held in the portfolio for

various time periods ranging from days to months and years depending on the researcher’s beliefs

about the duration of time it takes for the market to fully incorporate all relevant information into

the security valuation and to correct the initial mis-pricing. To assess the significance of the

valuation adjustment, a risk model is fitted to the time series of portfolio returns. We use the

standard four-factor asset pricing model (Fama-French 1993, Carhart 1997), allow for time-

varying risk factor loadings (Fama 1998), and include the Lewellen and Nagel (2006) correction

for high-frequency data estimation:


30

Retpt-RiskFreet = αp + ∑=

Q

q 1Qq∑

Τ

=0t

[βmkt,pqτ(RetMktt-τ-RiskFreet-τ) + βSMB,pqτSMBt-τ

+ βHML,pqτHMLt-τ + βUMD,pqτUMDt-τ] + εpt, where

(6)

qQ is a set of indicator variables equal to 1 when the rebalancing period is q and zero otherwise,

and the other variables are defined as previously. The intercept αp is the estimate of abnormal

return for portfolio p. It reflects the average return not explained by the risk profile of portfolio p,

and is interpreted as an abnormal return due to the attribute used to form portfolio p. If a

significant αp is found, it is said that the market initially did not correctly impound the value

implications of the signal contained in the information set used to form portfolio p. Under efficient

markets, no mispricing exists and the intercept αp should not differ from zero. Model 6

specification accounts for the varying portfolio risk associated with rebalancing (i.e., the firms in

the portfolio change every time new mergers are completed and their stock is added or removed

from the portfolio) and accommodates potential time-based variation in risk loadings (Ang and

Kristensen 2012). Model 6 also incorporates the Lewellen and Nagel (2006) correction for

estimation with high-frequency data (it includes both current and lagged risk factors).3

We compute our portfolio returns for various holding periods (e.g., our 3-year hold raw

portfolio returns are depicted in Figure 1) and estimate model 6. We present the results in Table 9.

For all alternative holding periods we examine (1, 2, 3, 4, and 5 years), the abnormal returns for

firms undertaking assimilation and fusion branding are not different from zero. This finding

indicates that the market correctly prices the choice of branding strategy undertaken by firms

pursuing fusion and assimilation branding strategies. The differential between these two portfolios

is also not significant at any holding period.

Interestingly, we observe significant negative abnormal stock returns to firms pursuing

business-as-usual branding in two-, three-, four-, and five-year holding periods. This finding

3 Lewellen and Nagel (2006) observed that although daily data allow for more precise estimates, non-synchronous prices (i.e., a delay in a response to common effects) can have a significant impact on the estimation of short-window risk covariates. As we use daily data, following Lewellen and Nagel (2006), we include both current and lagged risk factors in our model.


31

indicates the initial large negative reaction to business-as-usual merger announcements we

document in the event study is not complete. The post-merger negative adjustment we find implies

the market does not fully appreciate all negative consequences of a business-as-usual merger at the

time of the merger announcement and it takes a long time after merger completion to recognize

and correct the initial mis-valuation. The abnormal portfolio returns for business-as-usual-branded

mergers are significantly negative compared to fusion-branded mergers with two- to five-year

holding periods. The business-as-usual-branded mergers also significantly underperform

assimilation-branded mergers at three- and four-year holding periods.

To summarize, the negative post-M&A drift documented in prior literature appears to be

driven entirely by the business-as-usual-branded mergers. The pattern of our findings is consistent

with the stock market correctly pricing the implications of corporate branding for assimilation and

fusion branding strategies. We find that initially the market is overly optimistic (i.e., the initial

negative stock market reaction documented in tests of H1 is not sufficiently negative) about the

prospects of firms choosing business-as-usual branding and it devalues these mergers in the years

following merger completion.

Sensitivity analyses

We undertook a number of sensitivity analyses to assess the stability and reliability of our results

and found no evidence to question the findings we report. We tested alternative event windows

(e.g., from one to 20 days around announcement date) and alternative models for computing

abnormal returns (e.g., market model) in the event study analyses. We tested alternative

specifications of the inverse probability weighting and various additional controls in the regression

adjustment in our event study and operating performance treatment effects models (e.g., size

differential, profitability differential, strategic emphasis, marketing intensity, etc.) We examined

alternative specifications (e.g., constant vs. time-varying risk factor loadings) of the calendar time

portfolio models. We tested the sensitivity of our results to the inclusion of the three “brand new”

post-merger brand observations and undertook all tests including and excluding them. We also


32

undertook additional sensitivity tests on alternative samples described below. In all our tests we

found no differences with results we report.

Testing the influence of brand perceptions

One additional question we consider is the impact of perceptual brand characteristics of the

merging firms. While complete perceptual data is not available for all brands, we were able to use

Y&R BAV data (Mizik and Jacobson 2008) to obtain brand perceptions measures for 73 acquirers

(target data were available in only 10 cases). Of the 73 cases, 40 were assimilation, 17 were

business-as-usual, and 16 were fusion-branded M&As (i.e., the distribution of branding strategies

does not differ from that in our study sample). We replicated our event study using the sample of

73 deals and models augmented with brand perceptions data. These models also explicitly control

for the non-random data availability in the BAV database using a Heckman correction.4 We find

that while the included BAV measures are significant in the regression adjustment stage, the

overall pattern of our results holds in this smaller BAV sample: fusion has a significantly more

positive market reaction than assimilation and business-as-usual-branded mergers. Notably,

though, we find that including brand perceptions data and a control for attrition (for not being

covered in the BAV database) increases the magnitude and the statistical significance of the

estimate for fusion-branded announcements. The estimated average effect of market reaction to

fusion M&A announcements becomes positive and significant (3.766, p<.01) in this sample.

Testing temporal stability of results with a post-2006 M&A data sample

To assess the temporal stability of our findings we collected additional data on M&As announced

and completed after 2006. We identified M&A deals between two domestic public companies

announced in 2007-2017 in the SDC Platinum database. We used company press releases and

media reports to identify branding (name and symbol) used before and after a merger. When those

were unavailable, we relied on post-merger logos of target and acquirer and the target firm’s

4 In the first stage for Heckman estimation we ran a probit of “missing-from-BAV” indicator on observables, and found that coverage in the BAV database is predicted by acquirer size and its advertising intensity. We compute an Inverse Mills ratio and use it as an additional control in our regression adjustment stage. This approach is similar to our attrition modeling employed for testing H2 and H3.


33

website to classify branding strategy. We started with the largest deals (>$25B). Each M&A case

was discussed and was included into the data set only when there was a consensus on the branding

strategy classification among three experts evaluating it. At lower transaction values (<$5B), our

ability to reliably identify branding elements declined significantly and data collection was halted.

The procedure generated 178 deals (71 assimilation, 87 business-as-usual, 20 fusion). We

replicated the tests of our hypothesis on these data and found that our results hold. We find no

significant differences in the pattern of the immediate (event study) and delayed stock market

reaction (CTP) to post-M&A branding, and find that the operating performance patterns in this

post-2006 data sample do not differ from those we report.

Discussion

Our results show that corporate brand strategy in mergers is highly value-relevant and has

significant implications for post-merger operating performance. We find significant differences in

the immediate market reaction to merger announcements and in the post-M&A sales and cost

growth rates depending on the choice of the post-M&A branding. We also find a systematic post-

merger adjustment of firm valuation for the business-as-usual-branded M&As.

Consistent with our hypotheses, fusion branding exhibits significantly more positive value

implications than assimilation and business-as-usual branding. Interestingly, we find that fusion-

branded mergers do not generate immediate negative market reaction at the time of the merger

announcement, and we find no systematic negative future-term adjustment in the valuation of

these firms (although we find a positive post-merger drift, it is not significant). As such, our

findings for this group of firms differ significantly from findings reported in past research on

mergers in general and from the two other groupings of mergers we examine.

We find somewhat more negative (albeit not significantly so) immediate market reaction to

the business-as-usual than to assimilation branding. Our tests of post-M&A operating performance

help explain this finding: the sales growth rates are not significantly different, but the costs

increase significantly more for business-as-usual than for the assimilation-branded mergers.

Further, we also find a significant negative post-M&A drift for the business-as-usual-branded


34

mergers. That is, the initial negative market reaction is followed by an additional negative

valuation adjustment after the merger completion. As we do not find a significant drift for either

fusion or assimilation, the significant negative drift (valuation adjustment) for the business-as-

usual mergers suggests that the implications of the business-as-usual branding are not fully priced

at the time of the merger announcement (i.e., the market over-estimates the prospects of business-

as-usual mergers).

Our study makes several important contributions. Most importantly, our study highlights

significant implications of corporate branding for valuation and for the future operating

performance. Past research seeking to distinguish between the signaling and the demand-shifting

explanations of market reaction to corporate name changes tends to argue for the signaling role of

corporate name changes (e.g., Horsky and Swyngedouw 1987). This research, however, did not

examine the operating performance consequences directly. Our findings show that branding

strategy affects operating performance outcomes (sales and costs), that these effects are consistent

with the implications branding strategy has for the firm’s stakeholders (customers and employees),

and that the market recognizes them early on.

Past research has documented a negative post-M&A valuation adjustment for the merged

firms. We show that the negative post-M&A drift is primarily driven by the business-as-usual-

branded mergers. We find no market mis-valuation for the fusion and assimilation-branded

mergers. That is, investors appreciate the clarity of the assimilation and fusion strategies but

initially have difficulty in properly pricing (i.e., they systematically over-value) the business-as-

usual-branded mergers.

Our findings have important managerial implications. As Hsu et al. (2016) discuss,

companies may be rather myopic in their approach to branding as they view it in narrowly internal

and operational terms. As a consequence, they are rarely deliberate in selecting branding strategies

and often do so in an ad hoc manner (Ettenson and Knowles 2006). Our findings highlight the

importance of the post-merger branding. We show that corporate brands are valuable assets and


35

suggest that branding of a merger should be a result of careful analyses of the value merging

brands hold for their firms’ stakeholders.

Conclusion

Mergers are disruptive events with major organizational implications. Corporate brands are

valuable market-based assets that undergo deliberate and often dramatic transformation following

a merger. They can change their meaning, image, identity, personality, and values and, as a result,

can prompt customers and employees to reevaluate their relationships and commitments and

investors to change their expectations of the firm’s future. Corporate branding can help mitigate

some of the uncertainty caused by a merger by clarifying the intent of the merger to the customers,

employees, and investors. Prior research and anecdotal evidence suggest that branding decisions

are not given the careful consideration afforded to other aspects of merger transaction. It is

imperative to change current business practices. Marketing professionals should be a part of the

M&A planning and due-diligence process. Mergers should be viewed as relationships and not as

transactions. Branding as an important tool for facilitating post-M&A integration and for

managing relationships with key stakeholders—customers, employees, and investors—the three

key audiences whose ongoing loyalty largely determines the success or failure of a merger.


36

References Agrawal, Anup, Jeffrey F. Jaffe, and Gershon N. Mandelker (1992), “The Post-Merger Performance of

Acquiring Firms: A Re-examination of an Anomaly,” Journal of Finance, 47 (4), 1605-1621. Ahmed, Nabila, Ruth David, and Ed Hammond (2018), “Dealmakers See Slower 2019 as Stocks, Politics Drag

on M&A,” Bloomberg, December 20, 2018, 8:32 PM PST. Ailawadi, Kusum L., Donald R. Lehmann, and Scott A. Neslin (2003), “Revenue Premium as an Outcome

Measure of Brand Equity,” Journal of Marketing, 67 (4), 1-17. Andrade, Gregor, Mark Mitchell, and Erik Stafford (2001), “New Evidence and Perspectives on Mergers,”

Journal of Economic Perspectives, 15 (2), 103-120. Ang, Andrew and Dennis Kristensen (2012), “Testing Conditional Factor Models,” Journal of Financial

Economics, 106 (1), 132-156. Arellano, Manuel and Stephen Bond (1991), "Some Tests of Specification for Panel Data: Monte Carlo

Evidence and an Application to Employment Equations," Review of Economic Studies, 58 (2), 277-297. Bahadir, S. Cem, Sundar G. Bharadwaj, and Rajendra K. Srivastava (2008), “Financial Value of Brands in

Mergers and Acquisitions: Is Value in the Eye of the Beholder?” Journal of Marketing 72 (11), 49-64. Barth, Mary E., Michael Clement, George Foster, and Ron Kasznik (1998), “Brand Values and Capital Market

Valuation,” Review of Accounting Studies, 3 (1-2), 41-68. Becker, Sally A. and Darryl Flamer (1997), “Checking Out Customers' Needs after The Deal,” Mergers and

Acquisitions, 32 (3), 8-10. Bekier, Matt and Michael J. Shelton (2002), “The Top-Line Route to Premium Merger Values,” Mergers and

Acquisitions, 37 (6), 23. Bharadwaj, Sundar G., Kapil R. Tuli, and Andre Bonfrer (2011), "The Impact of Brand Quality on Shareholder

Wealth," Journal of Marketing, 75 (5), 88-104. Bommaraju, Raghu, Michael Ahearne, Zachary R. Hall, Seshadri Tirunillai, and Son K. Lam (2018), “The

Impact of Mergers and Acquisitions on the Sales Force,” Journal of Marketing Research, 55 (2), 254-264. Bottomley, Paul A. and Stephen J. S. Holden (2001), “Do We Really Know How Consumers Evaluate Brand

Extensions? Empirical Generalizations Based on Secondary Analysis of Eight Studies,” Journal of Marketing Research, 38 (4), 494-500.

Buil, Isabel, Sara Catalan, and Eva Martinez (2016), “The Importance of Corporate Brand Identity in Business Management: An Application to the UK Banking Sector,” Business Research Quarterly, 19(1), 3-12.

Busso, Matias, Justin McCrary, John DiNardo (2014), “New Evidence on the Finite Sample Properties of Propensity Score Matching and Reweighting Estimators,” Review of Economics and Statistics, 96 (5), 885-897.

Cao, Zixia and Alina Sorescu (2013), “Wedded Bliss or Tainted Love? Stock Market Reactions to the Introduction of Cobranded Products,” Marketing Science 32 (6), 939-959.

Capron, Laurence (1999), “The Long-Term Performance of Horizontal Acquisitions,” Strategic Management Journal, 20 (11), 987-1018.

Capron, Laurence and John Hulland (1999), “Redeployment of Brands, Sales Forces, and General Marketing Management Expertise Following Horizontal Acquisitions: A Resource-Based view,” Journal of Marketing, 63 (2), 41-54.

Carhart, Mark (1997), “On Persistence in Mutual Fund Performance,” Journal of Finance, 52 (1), 57-82. Cheng, Terence C., and Pravin K. Trivedi (2015), "Attrition Bias in Panel Data: A Sheep in Wolf's Clothing? A

Case Study Based on the Mabel Survey," Health Economics, 24 (9), 1101-1117. Corrado, Charles J. (1989), “A Nonparametric Test for Abnormal Security-Price Performance in Event

Studies,” Journal of Financial Economics, 23 (2), 385-395.


37

Cox, David R. (1972), "Regression Models and Life Tables (With Discussion)," Journal of the Royal Statistical Society, Series B 34, 187-220.

Daniel, Kent and Sheridan Titman (2003), “Market Reaction to Tangible and Intangible Information,” NBER, Working Paper 9743.

Eberhart, Allan C., William F. Maxwell, and Akhtar R. Siddique (2004) “An Examination of Long-Term Abnormal Stock Returns and Operating Performance Following R&D Increases,” Journal of Finance, 59 (2), 623-650.

Erdem, Tulin and Baohong Sun (2002), “An Empirical Investigation Of The Spillover Effects Of Advertising And Sales Promotions In Umbrella Branding,” Journal of Marketing Research, 39 (4), 408-420.

Ettenson, Richard and Jonathan Knowles (2006), “Merging the Brands and Branding the Merger,” MIT Sloan Management Review, 47 (4), 39-49.

Fama, Eugene (1998), “Market Efficiency, Long-term Returns, and Behavioral Finance,” Journal of Financial Economics, 49 (3), 283-306.

Fama, Eugene F., and Kenneth R. French (1993), “Common Risk Factors in The Returns on Stocks and Bonds,” Journal of Financial Economics, 33 (1), 3-56.

Fischer, Marc (2007), “Valuing Brand Assets: A Cost-Effective and Easy-to-Implement Measurement Approach,” MSI Working Paper Series, Report 07-107.

Gielens, Katrijn, Linda M. Van De Gucht, Jan-Benedict E. M. Steenkamp, et al. (2008), “Dancing with a Giant: The Effect of Wal-Mart's Entry into the United Kingdom on the Performance of European Retailers,” Journal of Marketing Research, 45 (5), 519-534.

Goldfarb, Avi, Qiang Lu, and Sridhar Moorthy (2009), “Measuring Brand Value in an Equilibrium Framework,” Marketing Science, 28 (1), 69–86.

Grambsch, Patricia M., and Terry M. Therneau (1994), "Proportional Hazards Tests and Diagnostics Based on Weighted Residuals," Biometrika 81 (3), 515-526.

Hambrick, Donald C., and Albert A. Cannella (1993), “Relative Standing: A Framework for Understanding Departures of Acquired Executives,” Academy of Management Journal, 36 (4), 733-762.

Han, Kyuhong, Vikas Mittal, Yan Zhang (2017), “Relative Strategic Emphasis and Firm-Idiosyncratic Risk: The Moderating Role of Relative Performance and Demand Instability,” Journal of Marketing, 81(4), 25-44

Harford, Jarrad, Mark Humphery-Jenner, and Ronan Powell (2012), “The Sources of Value Destruction in Acquisitions by Entrenched Managers,” Journal of Financial Economics, 106 (2), 247–261.

Harrison, Jeffrey S., Michael A. Hitt, Robert E. Hoskisson, R. Duane Ireland (2001), “Resource Complementarity in Business Combinations: Extending the Logic to Organizational Alliances,” Journal of Management, 27 (6), 679-690.

Healy, P., K. Palepu, and R. Ruback (1992), “Does Corporate Performance Improve after Mergers?” Journal of Financial Economics, 31 (1), 135–175.

Hirano, Keisuke, and Guido W. Imbens (2001), “Estimation of Causal Effects Using Propensity Score Weighting: An Application to Data on Right Heart Catheterization” Health Services and Outcomes Research Methodology, 2 (3–4), 259–78.

Hitsch, Günter J. (2006),"An Empirical Model of Optimal Dynamic Product Launch and Exit Under Demand Uncertainty," Marketing Science, 25(1), 25-50.

Homburg, Christian, and Matthias Bucerius (2005), “A Marketing Perspective on Mergers and Acquisitions: How Marketing Integration Affects Postmerger Performance,” Journal of Marketing, 69 (1), 95-113.

Horsky, Dan and Patrick Swyngedouw (1987), “Does It Pay to Change Your Company’s Name? A Stock Market Perspective,” Marketing Science, 6 (Fall), 320-34.


38

Hsu, Liwu, Susan Fournier, and Shuba Srinivasan (2016), “Brand Architecture Strategy and Firm Value: How Leveraging, Separating, and Distancing the Corporate Brand Affects Risk and Returns,” Journal of the Academy of Marketing Science, 44 (2), 261-280.

Hu, Yansong, and Christophe Van den Bulte (2014), "Nonmonotonic Status Effects in New Product Adoption," Marketing Science, 33 (4),. 509-533.

Imbens, Guido and Jeffrey Wooldridge (2009), “Recent Developments in the Econometrics of Program Evaluation,” Journal of Economic Literature, 47 (1), 5-86.

Imbens, Guido and Donald Rubin (2015). Causal Inference. Cambridge University Press. Jacobson, Robert and Natalie Mizik (2009), “Customer Satisfaction-Based Mispricing: Issues and

Misconceptions,” Marketing Science, 28 (5), 836-845 Jensen, Michael C. (1986), “Agency Costs of Free Cash Flow, Corporate Finance and Takeovers,” American

Economic Review, 76 (2), 323-329. Jensen, Michael C. (1993), “The Modern Industrial Revolution, Exit, and the Failure of Internal Control

Systems,” Journal of Finance, 48 (3), 831-880. Kalaignanam, Kartik and Cem S. Bahadir (2013) “Corporate Brand Name Changes and Business Restructuring:

Is The Relationship Complementary Or Substitutive?” Journal Of The Academy of Marketing Science, 41 (4), 456-472.

King, David R., Dan R. Dalton, Catherine M. Daily, Jeffrey G. Covin (2004), “Meta-Analyses of Post-Acquisition Performance: Indications of Unidentified Moderators,” Strategic Management Journal, 25 (2), 187-200.

Kroll, Mark, Peter Wright, Leslie Toombs, Hadley Leavell (1997), “Form of Control: A Critical Determinant of Acquisition Performance and CEO Rewards,” Strategic Management Journal, 18 (2), 85-96.

Krug, Jeffrey A. and Walt Shill (2008), “The Big Exit: Executive Churn in The Wake of M&As,” The Journal of Business Strategy, 29 (4), 15-21.

Lane, Vicki and Robert Jacobson (1995), “Stock Market Reactions to Brand Extension Announcements: The Effects of Brand Attitude and Familiarity,” Journal of Marketing, 59 (1), 63-77.

Lewellen, Jonathon and Stefan Nagel (2006), "The Conditional CAPM Does Not Explain Asset-Pricing Anomalies," Journal of Financial Economics, 82 (2), 289-314.

Li, K., N.R. Prabhala (2007), “Self-Selection Models in Corporate Finance.” In Eckbo, B.E. (Ed.), Handbook of Corporate Finance: Empirical Corporate Finance. Elsevier North Holland, Amsterdam, 37–86.

Li, Xiaoyang (2013), “Productivity, Restructuring, and the Gains from Takeovers,” Journal of Financial Economics, 109 (1), 250-271.

Lowrey, Tina M. and L. J. Shrum (2007), “Phonetic Symbolism and Brand Name Preference,” Journal of Consumer Research, 34 (3), 406-414.

MacKinlay, A. C. (1997), “Event Studies in Economics and Finance,” Journal of Economic Literature, 35 (1), 13-39.

McAlister, Leigh, Raji Srinivasan, MinChung Kim (2007), “Advertising, Research and Development, and Systematic Risk of the Firm,” Journal of Marketing, 71(1), 35-48.

Melnyk, Valentyna, Kristina Klein, and Franziska Voelckner (2012), “The Double-Edged Sword of Foreign Brand Names for Companies from Emerging Countries,” Journal of Marketing 76 (6), 21-37.

Mitchell, Mark L. and Erik Stafford (2000), “Managerial Decisions and Long-Term Stock Price Performance,” The Journal of Business, 73 (3), 287-329.

Mizik, Natalie and Robert Jacobson (2009), “Valuing Branded Businesses,” Journal of Marketing, 73 (6), 137-153.


39

Mulherin, J. Harold, and Audra L. Boone (2000), “Comparing Acquisitions and Divestitures,” Journal of Corporate Finance, 6 (2), 117–139.

Newmeyer, Casey, Vanitha Swaminathan, John Hulland (2016), “When Products and Brands Trade Hands: A Framework for Acquisition Success,” Journal of Marketing Theory and Practice, 24 (2), 129-146.

Peterson, Robert A. and Ivan Ross (1972), “How to Name New Brands,” Journal of Advertising Research, 12 (6), 29-34.

Rao, Vithala R., Manoj K. Agarwal, Denise Dahlhoff (2004), “How Is Manifest Branding Strategy Related to the Intangible Value of a Corporation?” Journal of Marketing, 68 (4), 126-141.

Rego, Lopo L., Matthew T. Billett, and Neil A. Morgan (2009), "Consumer-Based Brand Equity and Firm Risk," Journal of Marketing, 73 (6), 47-60.

Roll, Richard (1986), “The Hubris Hypothesis of Corporate Takeovers,” Journal of Business, 59 (2), 197-216. Rossi, Peter (2018), “Causal Inference in Marketing Applications,” in Mizik and Hanssens (Eds.) Handbook of

Marketing Analytics. Elgar Publishing. Siegel, Donald S., and Kenneth L. Simons (2010), “Assessing the Effects of Mergers and Acquisitions on Firm

Performance, Plant Productivity, and Workers: New Evidence from Matched Employer-Employee Data,” Strategic Management Journal, 31 (8), 903-916.

Sood, Sanjay and Kevin Lane Keller (2012), “The Effects of Brand Name Structure on Brand Extension Evaluations and Parent Brand Dilution,” Journal of Marketing Research, 49 (3), 373-382.

Sorescu, Alina, Rajesh K. Chandy, Jaideep C. Prabhu (2007), "Why Some Acquisitions Do Better than Others: Product Capital as a Driver of Long-term Stock Returns," Journal of Marketing Research, 44 (1), 57-72.

Sorescu, Alina, Venkatesh Shankar, Tarun Kushwaha (2007), “New Product Preannouncements and Shareholder Value: Don't Make Promises You Can't Keep,” Journal of Marketing Research 44 (3), 468-489.

Srinivasan, V., Chan Su Park, and Dae Ryun Chang (2005), “An Approach to the Measurement, Analysis, and Prediction of Brand Equity and Its Sources.” Management Science, 51 (9), 1433-48.

Swaminathan, Vanitha, Christopher Groening, Vikas Mittal, and Felipe Thomaz (2014) “How Achieving the Dual Goal of Customer Satisfaction and Efficiency in Mergers Affects a Firm’s Long-Term Financial Performance,” Journal of Service Research, 17 (2),182-194.

Swaminathan, Vanitha, Feisal Murshed, and John Hulland (2008) "Value Creation Following Merger and Acquisition Announcements: The Role of Strategic Emphasis Alignment," Journal of Marketing Research, 45 (1), 33-47.

Tavassoli, Nader T., Alina Sorescu, Rajesh Chandy (2014), “Employee-Based Brand Equity: Why Firms with Strong Brands Pay Their Executives Less,” Journals of Marketing Research, 51 (December), 676-690.

Thorbjornsen, Helge and Micael Dahlen (2011), “Customer Reactions to Acquirer-Dominant Mergers and Acquisitions,” International Journal of Research in Marketing 28 (4), 332-341.

Wooldridge, Jeffrey M. (2010). Econometric Analysis of Cross Section and Panel Data. 2nd ed. Cambridge, MA: MIT Press.

Wooldridge, Jeffrey M. (2015), “Control Function Methods in Applied Econometrics,” Journal of Human Resources, 50 (2), 420-45.

Zahra, Shaker A. and Gerard George (2002), “Absorptive Capacity: A Review, Reconceptualization, and Extension,” The Academy of Management Review, 27 (2), 185-203.

Zamudio, Cesar and Vanitha Swaminathan (2016), “The Dark Matter of Firm Equity: The Hidden Role of Employee-Based Brand Equity in Long-Term Stock Returns,” working paper, U. Pittsburgh.


40

Table 1. Branding Strategies in Mergers

Table 1A. Key Differences in M&A Branding Strategies

Financial Implications

Potential Strategic Rationale

Brand Equity Message Signal Immediate

Rebranding Cost

On-going Brand

Maintenance Cost

Likely Post-M&A Sales

Growth

Likely Post-M&A

Operating Costs

Assimilation Growing scale, consolidating market power

Fully discarded for one brand

Clear statement of dominance of one entity over the other

Strong commitment to integration via dominance

Medium Low Diminished Increased

Business-as-Usual

Diversification, expanding product and customer portfolio

Fully preserved for both brands Independence No clear signal

for integration Zero or Low High Neutral Increased

Fusion

Combination of the capabilities and cultures of the two companies

Modified to enhance

Unity, collaboration and respect for both entities

Strong commitment to integration as equal partners

High Low Neutral or Increased Diminished


41

Table 1B. Examples of Branding Strategies in Mergers Table 1B. Panel (a). Three Merger Branding Strategies in the Airline Industry

Firm 1 Branding Firm 2 Branding Resulting Branding

Assimilation Branding

Delta Airlines

Northwest Airlines

Delta Airlines

Business-as-Usual Branding

Fusion Branding

Table 1B. Panel (b). Examples of Fusion Branding in Mergers Firm 1 Branding Firm 2 Branding Resulting Branding

Mixture of Symbol and Name

Mixture of Names

Anheuser-Busch

InBev

Anheuser-Busch InBev

New Brand Entity


42

Table 2. Data Sample Characteristics

Table 2A. Sample Distribution across Industrial Groupings

Assimilation Business-As- Usual Fusion

Agriculture, Forestry, and Fishing 0 1 0 Mining 6 1 3 Manufacturing 59 19 13 Transportation, Communications, Electric,

Gas, and Sanitary Services 14 5 9 Wholesale Trade 0 0 1 Retail Trade 5 9 2 Finance, Insurance, and Real Estate 17 6 11 Services 17 8 4 TOTAL 118 49 43

Table 2B. Study Sample Coverage of M&A Transactions by Transaction Value Number of Transactions Total $ Value of Transactions ($b) Transactions Study Sample SDC Platinum Study Sample SDC Platinum over $25 billion 31 32 $1,564 $1,599 over $10 billion 69 85 $2,170 $2,423 over $1 billion 118 563 $2,310 $3,863 over $100 million 210 1748 $2,358 $4,265 Table 2C. Pre-Merger Characteristics of Merging Firms (n=210)

Mean S.E. 10th Pct. Median 90th Pct. Acquiring Firms Net Income ($M) 1701 195 22 530 5807 Sales ($M) 16964 1589 792 9527 43838 Assets ($M) 62333 10789 1226 12711 135792 Market Cap ($M) 40283 4300 1440 15108 114367 Book-to-Market Ratio .355 .017 .106 .322 .646 Target Firms Net Income ($M) 393 73 -47 33 1557 Sales ($M) 4906 628 97 777 14519 Assets ($M) 19267 4475 141 806 47570 Market Cap ($M) 9566 1206 153 1321 32351 Book-to-Market Ratio .394 .029 .076 .360 .741


43

Table 2D. Sample Characteristics by Corporate Branding Type (n=210) Assimilation Business-as-Usual Fusion Difference Mean S.E. Mean S.E. Mean S.E. p-value Brand Name Announced .110 .029 .163 .053 .349 .074 <.01 Horizontal Merger .712 .042 .408 .071 .442 .077 <.01 H of Brands Strategy .076 .025 .327 .068 .116 .049 <.01 Merger Premium 1.442 .060 1.268 .025 1.248 .041 .04 Ratio of Assets (Target/Acquirer) .326 .044 .379 .059 .544 .125 .09

Acquirer Size 8.818 .148 8.587 .201 9.298 .202 .07 Acquirer Profitability .072 .006 .044 .010 .057 .007 .03 Acquirer Ad Intensity .011 .002 .022 .006 .010 .004 .06 Target Size 6.664 .183 7.055 .229 7.734 .310 <.01 Target Profitability -.015 .020 .049 .016 .045 .016 .04 Target Ad Intensity .008 .003 .039 .011 .019 .013 <.01 Acquirer Industry HHI .144 .014 .231 .028 .164 .022 <.01 Acquirer Industry Growth 1.350 .073 1.417 .129 1.213 .044 .41 Target Industry HHI .110 .008 .128 .015 .122 .014 .49 Target Industry Growth 1.275 .059 1.168 .028 1.270 .110 .55 Table 2 Legend: Merger Characteristics

Brand Name Announced Merger announcement includes the brand name of the new firm Horizontal Merger The acquirer and target come from the same industrial grouping House of Brands Strategy The acquirer’s subsidiaries operate and market products under their own brand

names (e.g., P&G) Merger Premium The ratio of the offer price for the target and target firm’s stock price before the

merger announcement Ratio of Assets The ratio of the target’s to the acquirer’s assets Acquirer and Target Characteristics

Size Log of assets in the last financial report before the merger Profitability Net income divided by total assets in the last financial report before the merger Ad Intensity Advertising spending divided by total assets in the last financial report before the

merger Industry Characteristics Industry HHI Herfindahl-Hirschman Index at the time of the merger Industry Growth The average revenue growth rate in the industry in the three years prior to merger


44

Table 3. Classic Event Study Analysis Table 3A. Daily Abnormal Returns by Branding Strategy Group

All Firms Assml BaU Fusion Day AR p-value AR p-value AR p-value AR p-value -10 .04% .83 .23% .32 -.10% .76 -.34% .37 -9 .19% .23 -.08% .65 .47% .29 .64%** .04 -8 .09% .60 -.07% .76 .40% .22 .16% .64 -7 -.17% .29 -.16% .48 -.15% .69 -.24% .42 -6 .18% .31 .49%* .07 -.26% .44 -.15% .55 -5 .22% .19 .47% .06 .12% .64 -.35% .24 -4 .25% .16 .17% .47 .10% .74 .61% .14 -3 -.10% .56 -.05% .86 -.09% .73 -.25% .46 -2 -.14% .44 -.11% .66 -.41% .36 .08% .77 -1 -.76%*** .00 -.98%*** .01 -.58% .21 -.33% .31 +0 -1.26%*** .00 -1.63%*** .00 -1.34%* .08 -.16% .86 +1 -.12% .62 -.40% .21 .60% .25 -.16% .75 +2 -.02% .92 .20% .39 -.17% .71 -.43% .05 +3 -.02% .90 .00% 1.00 .35% .24 -.50% .18 +4 -.03% .87 .00% .99 .17% .65 -.32% .27 +5 -.04% .78 -.14% .44 .00% 1.00 .20% .59 +6 -.14% .35 -.27% .15 .08% .82 -.04% .92 +7 .16% .32 .17% .48 -.05% .82 .38% .36 +8 .17% .30 .07% .78 .51% .10 .06% .79 +9 -.14% .37 -.13% .59 -.17% .57 -.14% .53 +10 .05% .76 .01% .94 .41% .25 -.29% .28

Table 3B. Cumulative Abnormal Returns for [-1;0] window by Branding Strategy Group

Event Window

All Firms (N=210)

Returns by Corporate Branding Group

Abnormal Return Differences across Branding Group

Assiml

(N=118)

BaU

(N=49)

Fusion (N=43)

Fusion

vs. Assiml

Fusion vs.

BaU

BaU vs.

Assiml [-1,0] window

CAR -2.02*** -2.61*** -1.92** -.49 ΔCAR 2.13* 1.44 .69 Std Err .45 .59 .98 .96 Std Err 1.14 1.38 1.11 p-value .00 .00 .05 .61 p-value .06 .30 .54 Corrado θ -5.93 -5.46 -2.89 -1.11

Note: ∑=

+−−

=N

i

i

Ks

ttK

N 1

120

)(2

11θ , where ∑ ∑

+= =

+−

−−

=2

1 1

2

1

12

12 2111)(

t

t

N

ti

ttKNtt

Kst

t , and Kit is the

rank of the abnormal return of security i at the event time period t, N is the number of securities in the group, and θ is distributed standard normal. *** denotes significance at the 1% level, ** at 5%, and * at 10% level.


45

Table 4. Calculating Propensity Scores with a Multinomial Logit (n=210) Fusion to

Assimilation (base) Fusion to Business-As-Usual (base)

Business-As-Usual to Assimilation (base)

b S.E. b S.E. b S.E. Intercept -2.119* 1.321 -2.592* 1.445 .473 1.174 Merger Characteristics Horizontal Merger -1.531*** .431 .125 .476 -1.657*** .425 H of Brands Strategy -.014 .668 -1.593** .650 1.579*** .536 Name Announcement 1.120** .489 1.292** .577 -.172 .571 Acquirer Characteristics Acquirer Size -.118 .167 .370** .189 -.487*** .164 Target Characteristics Target Size .381*** .140 -.114 .161 .495*** .146 Target Ad Intensity 6.369 4.482 -1.399 3.301 7.769** 4.015 Log-likelihood -173.984 Note: Variable selection for the model specification comes from a stepwise regression with a maximum p=.10 threshold for inclusion. *** denotes significance at the 1% level, ** at 5%, and * at 10% level.

Table 5. Modeling Selection and Additional Influences in the Event Study Analysis (n=210) Table 5 Panel A. Inverse Probability Weighted Regression Adjustment Model of Abnormal Returns

Event Window

Potential Outcome Means Average Treatment Effects

Assiml (N=118)

BaU

(N=49)

Fusion (N=43)

Fusion

vs. Assiml

Fusion vs. BaU

BaU vs. Assiml

[-1,0] window CAR -2.193*** -3.05*** -0.209 ΔCAR 1.984** 2.842** -.858 S.E. .647 1.176 .713 S.E. .952 1.369 1.333 p-value .00 .009 .77 p-value .037 .038 .519 Table 5 Panel B. Regression Adjustment Results Assimilation Business-As-Usual Fusion b S.E. b S.E. b S.E. Horizontal Merger -3.153** 1.233 -3.127 2.312 1.388 1.597 H of Brands Strategy 2.615 2.076 4.071** 1.994 1.306 1.141 Name Announcement 2.464 1.911 -3.748* 2.25 .111 1.50 Merger Premium .943** .446 .721 4.807 -2.018 2.682 *** denotes significance at the 1% level, ** at 5%, and * at 10% level.


46

Table 6. Post-M&A Survival across Branding Strategies Table 6. Panel A. Model-Free Evidence: Survival Rates by Branding Strategy Type in the Ten Years after M&A Completion This table presents the fraction of firms within each strategy grouping that survive at the end of each year relative to the year when merger was completed (year 1). A firm is counted as ‘‘surviving’’ in a given year if it is listed in the CRSP database at the end of that year. We use returns data through the 2015 (our M&A sample covers 1997-2006), to ensure sample truncation issues do not affect our calculations for the ten post-M&A years depicted below.

Year after M&A 1 2 3 4 5 6 7 8 9 10 Assimilation 100.0% 99.2% 98.3% 94.1% 90.7% 89.0% 83.1% 80.5% 77.1% 74.6% Business-as-Usual 100.0% 100.0% 98.0% 93.9% 87.8% 83.7% 75.5% 67.3% 65.3% 61.2% Fusion 100.0% 100.0% 100.0% 97.7% 95.3% 90.7% 86.0% 83.7% 83.7% 79.1% Table 6. Panel B. Cox Proportional Hazard Model

VARIABLES Coefficient Estimate Standard Error Branding strategy: Acquisition 0.578* (0.313) Business-as-usual 0.879** (0.363) Fusion (base) Covariates: Acquirer size -0.542*** (0.110) Merger premium 0.221* (0.128) Horizontal merger -0.801*** (0.282) Acquirer Industry HHI -2.587** (1.216) Acquirer SIC dummies Yes Year effective dummies Yes Log-likelihood -381.9 AIC 783.7 BIC 817.2 Wald 𝜒𝜒2 p-value <.001 Observations 210

Note: Fusion branding is the base category. Stepwise regression (with threshold 0.1) used for variable selection. Standard errors clustered by effective year in parentheses. *** p<0.01, ** p<0.05, * p<0.1.

50.0%

55.0%

60.0%

65.0%

70.0%

75.0%

80.0%

85.0%

90.0%

95.0%

100.0%

year=1 2 3 4 5 6 7 8 9 10

Assimilation Business as Usual Fusion


47

Table 7. Post-M&A Operating Performance Changes This table presents the estimates of the IPWRA models for operating performance metrics in the ten years following M&A completion. Table 7A. Sales Growth (as Percent Change)

Time Period

Potential Outcome Means Average Treatment Effects Assiml (N=118)

BaU (N=49)

Fusion (N=43)

Fusion vs. Assiml

Fusion vs. BaU

BaU vs. Assiml

1 year Sales 1.441 -4.488 -3.264 ΔSales -4.705 1.225 -5.930 p-value (0.586) (0.328) (0.370) p-value (0.311) (0.834) (0.256) 2 year Sales 5.858 6.096 11.751*** ΔSales 5.893 5.656 0.237 p-value (0.175) (0.288) (0.003) p-value (0.323) (0.417) (0.973) 3 year Sales 7.135 13.588* 17.182*** ΔSales 10.047 3.594 6.452 p-value (0.131) (0.074) (0.000) p-value (0.137) (0.683) (0.460) 4 year Sales 8.459 18.803** 24.354*** ΔSales 15.896** 5.551 10.344 p-value (0.101) (0.032) (0.000) p-value (0.034) (0.566) (0.290) 5 year Sales 8.364 22.380** 28.470*** ΔSales 20.106** 6.090 14.016 p-value (0.153) (0.012) (0.000) p-value (0.024) (0.580) (0.174) 6 year Sales 11.177* 25.223*** 46.718*** ΔSales 35.541*** 21.495* 14.046 p-value (0.068) (0.005) (0.000) p-value (0.000) (0.059) (0.187) 7 year Sales 16.542** 37.444*** 59.293*** ΔSales 42.751*** 21.849* 20.903* p-value (0.012) (0.000) (0.000) p-value (0.000) (0.084) (0.078) 8 year Sales 25.657*** 27.442* 66.120*** ΔSales 40.463*** 38.677** 1.786 p-value (0.000) (0.058) (0.000) p-value (0.000) (0.018) (0.909) 9 year Sales 25.733*** 40.620*** 68.909*** ΔSales 43.176*** 28.289* 14.887 p-value (0.001) (0.001) (0.000) p-value (0.000) (0.064) (0.313) 10 year Sales 29.066*** 33.578** 71.568*** ΔSales 42.502*** 37.990* 4.512 p-value (0.000) (0.030) (0.000) p-value (0.002) (0.052) (0.802) *** denotes significance at the 1% level, ** at 5%, and * at 10% level. Table 7A2. Regression Adjustments, Year 10 Post-Merger Sales Growth Assimilation Business-As-Usual Fusion b p-value b p-value b p-value Intercept 57.673 (0.275) 219.410 (0.357) 4.569 (0.976) Acquirer Characteristics Acquirer Size -7.015 (0.204) -12.961 (0.577) 23.536** (0.032) Acquirer Profitability 137.507 (0.300) -162.613 (0.476) -138.349 (0.503) Acquirer Ad Intensity 92.358 (0.886) -119.756 (0.724) -800.875** (0.025) Target Characteristics Target Size -0.260 (0.955) 0.538 (0.937) -14.512*** (0.001) Target Profitability -48.466* (0.081) 100.129 (0.533) -138.222** (0.029) Target Ad Intensity 201.640 (0.431) -249.968 (0.188) 409.688*** (0.004) Industry Characteristics Industry HHI -29.531 (0.554) -6.671 (0.932) -104.082* (0.079) Growth Rate 18.699 (0.129) -21.619** (0.037) 7.673 (0.539) Attrition hazard 8.526 (0.774) -62.074 (0.162) -54.082 (0.268) R2 0.11 0.21 0.63 *** denotes significance at the 1% level, ** at 5%, and * at 10% level.


48

Table 7B. Operating Expenses Growth (as Percent Change)

Time Period

Potential Outcome Means Average Treatment Effects Assiml (N=118)

BaU (N=49)

Fusion (N=43)

Fusion vs. Assiml

Fusion vs. BaU

BaU vs. Assiml

1 year Opex 0.872 -0.353 -1.740** ΔOpex -2.612** -1.387 -1.225 p-value (0.301) (0.723) (0.047) p-value (0.032) (0.286) (0.341) 2 year Opex 1.874 1.299 -1.728 ΔOpex -3.602** -3.027** -0.575 p-value (0.121) (0.165) (0.112) p-value (0.022) (0.040) (0.712) 3 year Opex 2.767** 2.683** 0.210 ΔOpex -2.557 -2.473 -0.084 p-value (0.040) (0.021) (0.875) p-value (0.178) (0.152) (0.962) 4 year Opex 2.845** 3.255** -0.154 ΔOpex -2.999 -3.409* 0.410 p-value (0.035) (0.012) (0.913) p-value (0.114) (0.069) (0.825) 5 year Opex 3.039** 2.877*** -0.624 ΔOpex -3.663 -3.501 -0.163 p-value (0.026) (0.005) (0.746) p-value (0.107) (0.105) (0.923) 6 year Opex 1.525 1.943** -1.508 ΔOpex -3.033* -3.451** 0.418 p-value (0.262) (0.038) (0.245) p-value (0.100) (0.033) (0.790) 7 year Opex 0.495 2.316*** -1.359 ΔOpex -1.854 -3.675*** 1.821 p-value (0.713) (0.002) (0.255) p-value (0.299) (0.009) (0.246) 8 year Opex -0.434 4.544*** -1.382 ΔOpex -0.947 -5.925*** 4.978** p-value (0.727) (0.003) (0.210) p-value (0.563) (0.002) (0.021) 9 year Opex 0.744 6.844*** -1.066 ΔOpex -1.810 -7.910*** 6.100* p-value (0.658) (0.006) (0.380) p-value (0.358) (0.005) (0.055) 10 year Opex 0.931 4.210** -0.446 ΔOpex -1.377 -4.656** 3.279 p-value (0.541) (0.016) (0.611) p-value (0.411) (0.022) (0.177) *** denotes significance at the 1% level, ** at 5%, and * at 10% level. Table 7B2. Regression Adjustments, Year 10 Post-Merger Operating Expenses Growth Assimilation Business-As-Usual Fusion b p-value b p-value b p-value Intercept 1.615 (0.873) 29.912* (0.098) -26.401*** (0.004) Acquirer Characteristics Acquirer Size 0.605 (0.434) -1.138 (0.575) 3.284*** (0.000) Acquirer Profitability 28.121 (0.170) -20.112 (0.292) -1.605 (0.930) Acquirer Ad Intensity -203.092*** (0.001) 73.638** (0.032) -11.930 (0.770) Target Characteristics Target Size -0.594 (0.476) -1.920** (0.011) -0.702 (0.248) Target Profitability 8.011 (0.136) -21.768 (0.364) -10.700* (0.063) Target Ad Intensity 0.478 (0.988) -36.750 (0.146) -47.302*** (0.005) Industry Characteristics Industry HHI 17.169* (0.096) -13.557 (0.141) 17.750*** (0.000) Growth Rate 0.663 (0.806) 2.549* (0.066) -2.483 (0.154) Attrition hazard -11.590*** (0.000) -3.901 (0.492) 4.918** (0.020) R2 0.27 0.39 0.76 *** denotes significance at the 1% level, ** at 5%, and * at 10% level.


49

Table 8. Risk Profile of the Branding Portfolios This table presents the measures of total, systematic, and idiosyncratic risk for Assimilation, Business-as-Usual, and Fusion portfolios. Total risk is measured as the variance of portfolio return. Systematic risk is measured as the market beta of the portfolio. Idiosyncratic risk is measured as the standard deviation of the residuals from the regression of portfolio returns on the time-varying four-factors of Carhart (1997) asset pricing model. Horizon All Firms Assimilation Business-as-Usual Fusion 3 Months Total Risk .00051 .00054 .00062 .00053 Systematic Risk 1.05339 1.08437 1.04144 1.08609 Idiosyncratic Risk .00972 .01248 .01597 .01496 6 Months Total Risk .00058 .00056 .00060 .00053 Systematic Risk 1.09160 1.08926 1.07780 1.13067 Idiosyncratic Risk .00708 .00945 .01430 .01391 1 Year Total Risk .00696 .00065 .00053 .00046 Systematic Risk 1.07953 1.09415 1.04308 1.10682 Idiosyncratic Risk .00534 .00741 .00970 .00965 2 Year Total Risk .00098 .00095 .00047 .00058 Systematic Risk 1.06929 1.09946 1.01871 1.09847 Idiosyncratic Risk .00536 .00789 .00708 .0088 3 Year Total Risk .00120 .00110 .00050 .00067 Systematic Risk 1.06589 1.10332 1.01643 1.09623 Idiosyncratic Risk .00425 .00577 .00662 .00781 5 Year Total Risk .00141 .00124 .00074 .00071 Systematic Risk 1.06911 1.09563 1.02006 1.12428 Idiosyncratic Risk .00362 .00503 .00672 .00611


50

Table 9. Calendar Time Portfolio Returns This table presents the results of Hypothesis 2 tests using the calendar-time portfolio approach with time-varying risk factor loadings and high-frequency data correction. The data are presented as % daily abnormal returns estimated using model [6]. ** denotes significance at the 5% level; * denotes significance at the 10% level.

Event Window All Firms (N=210)

Returns by Corporate Branding Group Abnormal Return Differences across Branding Group

Assimilation

(N=118)

BaU

(N=49)

Fusion (N=43)

Fusion vs. Assiml

Fusion vs. BaU

BaU vs. Assiml

1 Year Alpha (%) -.0067 .0022 -.0221 -.0084 ΔAlpha -.0105 .0138 -.0243 S.E. .0109 .0148 .0242 .0212 S.E. .0269 .0322 .0281 p-value .54 .88 .36 .69 p-value .70 .67 .39 2 Years Alpha (%) -.0049 .0034 -.0338** .0074 ΔAlpha .0040 .0412* -.0372* S.E. .0084 .0112 .0176 .0150 S.E. .0198 .0231 .0209 p-value .56 .76 .05 .62 p-value .84 .07 .07 3 Years Alpha (%) -.0066 -.0012 -.0378*** .0092 ΔAlpha .0104 .0470*** -.0365** S.E. .0072 .0096 .0143 .0130 S.E. .0170 .0193 .0175 p-value .36 .89 <.01 .48 p-value .54 .01 .04 4 Years Alpha (%) -.0096 -.0037 -.0440*** .0104 ΔAlpha .0142 .0544*** -.0403*** S.E. .0052 .0083 .0130 .0113 S.E. .0147 .0172 .0154 p-value .11 .65 <.01 .36 p-value .34 <.01 .01 5 Years Alpha (%) -.0088 -.0063 -.0273** .0045 ΔAlpha .0107 .0318** -.0211 S.E. .0056 .0073 .0118 .0100 S.E. .0129 .0155 .0138 p-value .12 .39 .02 .65 p-value .41 .04 .13 *** denotes significance at the 1% level, ** at 5%, and * at 10% level.


51

Figure 1. Raw Calendar Portfolio Returns for Branding Portfolios This figure presents raw calendar portfolio returns (3-year post-merger hold) for the three corporate branding strategies. The shaded area of the chart on the left highlights the initial 18-month period when we have a small number of equities in the portfolios. The shaded area on the right represents the end of our merger data sample; that is, new mergers are not entering our portfolios past 2006.

Figure 2. Post-Merger Buy-and-Hold Abnormal Returns This figure presents the average buy-and-hold abnormal returns for the three branding strategies for the 3-year period beginning 3 days after the merger completion date. The abnormal returns are computed as a difference between the raw return for the equity and the return accrued to Fama and French factors of the same period.

$-

$500

$1,000

$1,500

$2,000

$2,500

Mar

97

Mar

98

Mar

99

Mar

00

Mar

01

Mar

02

Mar

03

Mar

04

Mar

05

Mar

06

Mar

07

Mar

08

Assimilation Business-as-Usual Fusion

Fusion

Assimilation

Business-as-Usual

-30%

-25%

-20%

-15%

-10%

-5%

0%

5%

10%

15%

Year 0 Year 1 Year 2 Year 3

Assimilation Business-as-Usual Fusion

Business-as-Usual

Assimilation

Fusion


Documents

Branding a Merger: Implications for Merger Valuation and ...staff.washington.edu/epavlov/images/Branding a merger.pdfname and symbol), business-as-usual (both firms continue to operate