Calculating a bid premium

On March 20, 2016, U.S. paint retailer Sherwin-Williams Co. announced that it had signed
an agreement to acquire another U.S. paint maker, Valspar Corp., in an $11.3 billion deal.
Sherwin-Williams offered Valspar shareholders $113 per share in cash. Sherwin-Williams
also agreed to assume approximately $2 billion of Valspar’s debt (which you can ignore for
the purposes of this assignment).
In this question, you are going to calculate the bid premium paid by Sherwin-Williams
for Valspar. The first step in the process is determining what the appropriate “bid window”
period is. Should the calculation start from the day the takeover was announced (March
20th), or an earlier date? Did the market anticipate the announcement? If so, when did the
price begin to reflect the probability of a takeover bid? [Note: You can check news sources
such as Factiva to help you determine whether the market anticipated the announcement.]
Should the calculation end on the announcement date or sometime later, such as the completion date? There are no single “correct” answers to these questions, so you will need to
use your best judgment and justify your choice of a window period.
The second step in the process is to compute the expected and abnormal returns on
Valspar’s stock during the bid window. First, you will need to run a regression prior to your
bid window to estimate the α and β for Valspar’s stock. I would suggest using three years
1
of historical return data to run this regression. The regression you need to run is:
rV alspar − rf = α + β(rMarket − rf ) ,
where rV alspar is the daily or monthly return on Valspar’s stock, rf is the risk-free rate, and
rMarket is the return on the market portfolio. You can obtain historical data on Valspar’s
stock returns from any one of a number of sources (Bloomberg, Yahoo!, etc.). In general, you can obtain historical data on the risk-free rate and the market return from Ken
French’s website (click on “Data Library,” then “Fama/French Factors” for monthly returns
or “Fama/French Factors [Daily]” for daily returns.) However, you can also use another
source if you’d like. Then you’ll need to run the regression. One approach (which I generally
like) might be to run the regression using monthly returns for the 36 months preceding the
start of your bid window.
Once you’ve run the regression, the next step is to take the α and β estimates from your
regression and use them to compute the expected and abnormal returns on Valspar’s stock
during the window period. Here, you can use the procedure outlined in the lecture notes.
Let’s use March as an example. For March, you would take the actual excess return on
Valspar’s stock and then subtract (α + β*the actual excess market return in March). That
would give you Valspar’s abnormal return for March. You would do this for each month
during your bid window period and then sum up the abnormal return across the entire bid
window period. That would be your total bid premium.
As you are computing your bid premiums, be careful not to mix-and-match your benchmarks and/or return frequencies. For example, don’t estimate the model using monthly
returns and then compute daily abnormal returns during the bid window. Likewise, don’t
use the S&P 500 to estimate the model and then use the Dow Jones Industrial Average to
compute abnormal returns during the bid window. Good luck!
2 Merger arbitrage [30 points]
Let’s continue analyzing the Valspar deal. Valspar’s stock opened at $104.95 on Monday,
March 21st (the deal was announced on Sunday), after closing at $83.83 on the last trading
day prior to the announcement. Sherwin-Williams’ stock opened flat, dropped over the next
few days, but then recovered fully.
The Valspar deal also included a twist that was quite novel in the world of takeovers. In
2
particular, the merger agreement stated that Valspar shareholders were entitled to receive
$113 per share in cash conditional on antitrust-related asset divestitures being less than $650
million (based on 2015 revenues). If antitrust regulators required the firms to sell assets that
accounted for more than $650 million in 2015 revenues, Valspar’s shareholders would only
receive $105 per share. Furthermore, if regulators required the firm to sell assets accounting for more than $1.5 billion in revenues, both parties could simply walk away from the deal.
Just after the deal was announced, the Wall Street Journal reported that the market
believed there is an 80% probability that antitrust regulators will demand less than $650
million in divestments, a 15% probability that regulators will demand more than $650 million
but less than $1.5 billion in divestments, and a 5% probability that regulators will demand
more than $1.5 billion in divestments.
(a) Compute a merger arbitrageur’s expected return from investing in this deal under
the assumption that the arbitrageur has no informational advantage over the rest of the market. You may assume that the annual cost of borrowing for the arbitrageur is 3% (which may
or may not be necessary for answering the question). You may also assume that if the deal
is completed, it will be completed exactly 6 months after the announcement date. Finally,
you may assume that the risk-free rate is zero. [10 points]