Counterpoint: Groupon is not a success
•
by Erin Griffith (http://fortune.com/2015/03/20/groupon-success/)
The company must overcome a bad business model.
Earlier this year, my esteemed colleague Dan Primack argued that Groupon GRPN 1.31% , the
daily deals company which is largely viewed as a failure, should not be viewed as such. It was a
great contrarian argument, except that it was wrong.
Dan’s argument boiled down to the following points, which I’ll address in order:
1.) Groupon is not a failure because it has a $4.9 billion market cap, which is more than it was
worth as a private company. (Its market cap has risen to $5.3 billion since his story.)
2.) Groupon is not a failure because it has steadily grown since its IPO, with plenty of cash on
hand and no debt, signals of a healthy company.
3.) Groupon was a victim of high expectations, and it’s not the company’s fault when it failed to
live up to them.
$5 billion company
Indeed, Groupon is worth more than it was ever valued as a private company. But barely.
Groupon’s last round of private funding valued it at $4.75 billion. That was in 2010. When Dan
published his story, Groupon was worth $4.9 billion. A 3% increase is not much of a return for
investors in five years, especially for a high-growth startup.
What’s worse, Groupon could have been worth more. In 2010, Google GOOG 0.80% offered to
acquire the company for $6 billion. After rejecting that deal, then-CEO Andrew Mason declared,
“Like, okay, we’re the best company in the world.” Today it’s clear that not many people believe
Groupon is the best company in the world, or that it’s worth $6 billion.
But the real reason a $5 billion valuation does not make Groupon a success is because anyone
who bought into Groupon’s growth story at IPO got burned. Groupon went public at a valuation
of $13 billion, trading up to $19.5 billion on its first day. Shortly after, the stock price cratered,
and it has been worth half of its IPO valuation (or less) ever since.
Even insiders got burned, since Groupon’s shares collapsed before the 90-day lock-up period
was over, meaning they couldn’t sell while their stock lost more than half of its value. Any
employee that exercised shares in the IPO likely owed more in taxes than their stock was worth.
Anyone who bought low has yet to see much growth either. Over the last three years, Groupon’s
stock has ended up relatively flat. It’s hard to call that a success.
Growth
I’m as surprised as Dan that people still spend $7.6 billion a year with Groupon. (That’s gross
sales—the company’s annual revenue was $3.2 billion last year.) Groupon is growing: It grew
revenue by 24% last year. Revenue is expected to grow 11% this year. But Groupon has reported
a net income loss for each year it’s been public, including last year, when it lost $73 million.
That’s not likely to change, because Groupon is a company that has to overcome a bad business
model.
By now we know the daily deals business model is a novelty. Even Andrew Mason (who was
ousted in 2013) admits the model stinks. Last month he told the Associated Press that Groupon
was a “stupid, boring idea that just happened to resonate.”
Groupon’s revenue from daily deals has been relatively flat since 2011. To diversify, Groupon
launched “Groupon Goods,” where it sells things like iPhones and jelly beans at a discount. This
business has grown to represent the majority of Groupon’s income, but the margins are
significantly smaller: 88% for daily deals versus 20% for goods.
The problem with both of Groupon’s businesses are the barriers to entry, or lack thereof. With
daily deals, anyone could buy a “build your own Groupon” software package (from a company
called Groupon Clone, no less) and start selling coupons with cute emails. In the height of
Groupon-mania, many did just that, driving Groupon’s customer acquisition costs sky-high and
forcing Groupon on an aggressive acquisition spree to take out regional competitors. At the
height of the boom there were hundreds of daily deals startups. Groupon bought more than 30 of
them. Now Amazon is moving in with its “Amazon Local” competitor. And Groupon Goods is
already competing with Amazon AMZN 0.58% , and every other online discounter.
To recap: flat growth in Groupon’s core business and some growth in Groupon’s low-margin
business. This is why Groupon’s stock has been uninspired for the past three years. Today, 14
analysts rate Groupon as a “Hold,” and only nine rate it “Buy” or “Strong Buy.”
High Expectations
Dan believes Groupon was a victim of media hype. I disagree. While no company can fully
control the media’s frenzied hype machine, Groupon poured fuel onto the fire. For example, the
aforementioned “greatest company in the world” line. Others examples include appearing on the
cover of Forbes with the headline “The fastest growing company ever” and a talk Mason gave on
how to “get super rich.” Mason did not tamp down expectations after a report that its IPO would
be worth $30 billion. Rather, he wrote detailed memo full of hype and promise about Groupon’s
future during the company’s quiet period, which was leaked (some believe on purpose) and got
him in trouble with the SEC.
Groupon was on such a crazy train ride that it didn’t have enough time to prove its model
actually worked before going public. This is why it ran into accounting issues like not
anticipating a high refund rate on big-ticket travel bookings, which caused it to restate its
earnings and sent its shares into a downward spiral. Some reports note that many of Groupon’s
investors and managers though it was too early for Groupon to go public.
After talk of a $30 billion IPO, Groupon’s brand is tainted so badly that it will be very difficult to
recover. It doesn’t matter that Groupon is growing (modestly) or able to do $3.2 billion in annual
revenue. Is Groupon a company that talented people want to work for? Is Groupon a company
exciting startups want to sell themselves to? I would argue no to each.
The hype factor, which again, was fueled by the company and its investors just as much as the
press, is the reason I must address Dan’s headline, which implores: guys, come on, let’s all stop
laughing at Groupon. I see no reason not to think of Groupon when I need a good chuckle, if
only for the following seven reasons:
1.) Mason’s cheeky resignation letter (“I was fired today. If you’re wondering why… you
haven’t been paying attention”). Actually, laugh at any of Mason’s goofball antics.
2.) Mason’s business rock album, “Hardly Working,” which is funny on purpose.
3.) This memo, which trashes the financial press for being critical of Groupon’s numerous
accounting errors and challenges.
4.) This time Groupon attacked a critical reporter for her “nastygram.”
5.) This time Groupon thought the Earth was 400 years old.
6.) I’m even able to laugh at the fact that, in order to access Groupon’s investor relations page, I
had to sign up for its daily deals blasts again, and the first deal I got was for a yoga studio I
frequent. I had no choice but to buy the coupon (it was a good deal!), thus indirectly supporting
my opponent’s argument. Dan had this to say:
7.) Lastly, after a few more weeks of irrelevant deals in my inbox, I got an offer for discount
PajamaJeans, which are always hilarious.
Counterpoint: Groupon is not a success
•
by Erin Griffith (http://fortune.com/2015/03/20/groupon-success/)
The company must overcome a bad business model.
Earlier this year, my esteemed colleague Dan Primack argued that Groupon GRPN 1.31% , the
daily deals company which is largely viewed as a failure, should not be viewed as such. It was a
great contrarian argument, except that it was wrong.
Dan’s argument boiled down to the following points, which I’ll address in order:
1.) Groupon is not a failure because it has a $4.9 billion market cap, which is more than it was
worth as a private company. (Its market cap has risen to $5.3 billion since his story.)
2.) Groupon is not a failure because it has steadily grown since its IPO, with plenty of cash on
hand and no debt, signals of a healthy company.
3.) Groupon was a victim of high expectations, and it’s not the company’s fault when it failed to
live up to them.
$5 billion company
Indeed, Groupon is worth more than it was ever valued as a private company. But barely.
Groupon’s last round of private funding valued it at $4.75 billion. That was in 2010. When Dan
published his story, Groupon was worth $4.9 billion. A 3% increase is not much of a return for
investors in five years, especially for a high-growth startup.
What’s worse, Groupon could have been worth more. In 2010, Google GOOG 0.80% offered to
acquire the company for $6 billion. After rejecting that deal, then-CEO Andrew Mason declared,
“Like, okay, we’re the best company in the world.” Today it’s clear that not many people believe
Groupon is the best company in the world, or that it’s worth $6 billion.
But the real reason a $5 billion valuation does not make Groupon a success is because anyone
who bought into Groupon’s growth story at IPO got burned. Groupon went public at a valuation
of $13 billion, trading up to $19.5 billion on its first day. Shortly after, the stock price cratered,
and it has been worth half of its IPO valuation (or less) ever since.
Even insiders got burned, since Groupon’s shares collapsed before the 90-day lock-up period
was over, meaning they couldn’t sell while their stock lost more than half of its value. Any
employee that exercised shares in the IPO likely owed more in taxes than their stock was worth.
Anyone who bought low has yet to see much growth either. Over the last three years, Groupon’s
stock has ended up relatively flat. It’s hard to call that a success.
Growth
I’m as surprised as Dan that people still spend $7.6 billion a year with Groupon. (That’s gross
sales—the company’s annual revenue was $3.2 billion last year.) Groupon is growing: It grew
revenue by 24% last year. Revenue is expected to grow 11% this year. But Groupon has reported
a net income loss for each year it’s been public, including last year, when it lost $73 million.
That’s not likely to change, because Groupon is a company that has to overcome a bad business
model.
By now we know the daily deals business model is a novelty. Even Andrew Mason (who was
ousted in 2013) admits the model stinks. Last month he told the Associated Press that Groupon
was a “stupid, boring idea that just happened to resonate.”
Groupon’s revenue from daily deals has been relatively flat since 2011. To diversify, Groupon
launched “Groupon Goods,” where it sells things like iPhones and jelly beans at a discount. This
business has grown to represent the majority of Groupon’s income, but the margins are
significantly smaller: 88% for daily deals versus 20% for goods.
The problem with both of Groupon’s businesses are the barriers to entry, or lack thereof. With
daily deals, anyone could buy a “build your own Groupon” software package (from a company
called Groupon Clone, no less) and start selling coupons with cute emails. In the height of
Groupon-mania, many did just that, driving Groupon’s customer acquisition costs sky-high and
forcing Groupon on an aggressive acquisition spree to take out regional competitors. At the
height of the boom there were hundreds of daily deals startups. Groupon bought more than 30 of
them. Now Amazon is moving in with its “Amazon Local” competitor. And Groupon Goods is
already competing with Amazon AMZN 0.58% , and every other online discounter.
To recap: flat growth in Groupon’s core business and some growth in Groupon’s low-margin
business. This is why Groupon’s stock has been uninspired for the past three years. Today, 14
analysts rate Groupon as a “Hold,” and only nine rate it “Buy” or “Strong Buy.”
High Expectations
Dan believes Groupon was a victim of media hype. I disagree. While no company can fully
control the media’s frenzied hype machine, Groupon poured fuel onto the fire. For example, the
aforementioned “greatest company in the world” line. Others examples include appearing on the
cover of Forbes with the headline “The fastest growing company ever” and a talk Mason gave on
how to “get super rich.” Mason did not tamp down expectations after a report that its IPO would
be worth $30 billion. Rather, he wrote detailed memo full of hype and promise about Groupon’s
future during the company’s quiet period, which was leaked (some believe on purpose) and got
him in trouble with the SEC.
Groupon was on such a crazy train ride that it didn’t have enough time to prove its model
actually worked before going public. This is why it ran into accounting issues like not
anticipating a high refund rate on big-ticket travel bookings, which caused it to restate its
earnings and sent its shares into a downward spiral. Some reports note that many of Groupon’s
investors and managers though it was too early for Groupon to go public.
After talk of a $30 billion IPO, Groupon’s brand is tainted so badly that it will be very difficult to
recover. It doesn’t matter that Groupon is growing (modestly) or able to do $3.2 billion in annual
revenue. Is Groupon a company that talented people want to work for? Is Groupon a company
exciting startups want to sell themselves to? I would argue no to each.
The hype factor, which again, was fueled by the company and its investors just as much as the
press, is the reason I must address Dan’s headline, which implores: guys, come on, let’s all stop
laughing at Groupon. I see no reason not to think of Groupon when I need a good chuckle, if
only for the following seven reasons:
1.) Mason’s cheeky resignation letter (“I was fired today. If you’re wondering why… you
haven’t been paying attention”). Actually, laugh at any of Mason’s goofball antics.
2.) Mason’s business rock album, “Hardly Working,” which is funny on purpose.
3.) This memo, which trashes the financial press for being critical of Groupon’s numerous
accounting errors and challenges.
4.) This time Groupon attacked a critical reporter for her “nastygram.”
5.) This time Groupon thought the Earth was 400 years old.
6.) I’m even able to laugh at the fact that, in order to access Groupon’s investor relations page, I
had to sign up for its daily deals blasts again, and the first deal I got was for a yoga studio I
frequent. I had no choice but to buy the coupon (it was a good deal!), thus indirectly supporting
my opponent’s argument. Dan had this to say:
7.) Lastly, after a few more weeks of irrelevant deals in my inbox, I got an offer for discount
PajamaJeans, which are always hilarious.
This estimate is even lower, but the range is even greater (with fewer observations), but again a big overlap with the other beta estimate ranges.
So the conclusion from the historical regression estimates is that the beta (for the past) is around 1 to 1.5, with a fair degree of stability. However, we'd still have to figure out if the future is likely to be similar to the past.
If we believe the Groupon business is going to remain "techy" in the future, we'd stick with this "high" beta estimate. On the other hand, if we believe the Groupon coupon business is likely to be countercyclical in the
future, we would go with a lower beta estimate.
Ultimately, we see that the final estimate still depends on our evaluation of Groupon's business. Data helps us come up with estimates, but in the final analysis, we have to apply economic analysis and our own common
sense.I could simply have computed the ratio of the Covariance between the returns on Groupon and the S&P to the variance of S&P returns and I would have got the same answer much more easily. But if I did that, I would not
know how much reliance to place on my computed estimate. With a regression, I can say (assuming that the returns are normally distributed), that the true underlying beta is between 1.018 and 2.063 with a probability of
95%. This gives me quite a bit of certainty that the true beta is greater than one!
Now I went ahead and regressed weekly Groupon returns on S&P weekly returns from Nov. 7, 2011 to April 18, 2016 and found the following:
Regression Statistics
Multiple R
R Square
Adjusted R
Square
Standard
Error
Observations
ANOVA
Regression
Residual
Total
Intercept
Return
Multiple R
0.363181
0.131901
R Square
Adjusted R
Square
Standard
Error
0.128143
Observations
0.089594
233
df
Regression Statistics
SS
MS
F
1 0.281739 0.281739 35.0986
1.854255 0.008027
2.135994
231
232
This suggests a higher beta; on the other hand, this is for the period Nov. 4, 2011 to April 18, 2016. Perhaps the beta has been decreasing over time, so that a beta computed using only the last year would have a lower beta.
So I repeated the regression using the same time period as the daily regression (of course now I would have much fewer data observations). My results were as follows:
Standard
Coefficients Error t Stat P-value Lower 95%
-0.00734 0.005921 -1.23929 0.216495
1.935643
-0.019
0.326724 5.924407 1.13E-08
1.291904
0.406737
0.165435
0.149071
Significance
F
1.13E-08
0.087252
53
Upper
95%
0.004328
2.579382ANOVA
Regression
Residual
Total
Intercept
X Variable 1
Multiple R
R Square
Adjusted R
Square
Standard
Error
Observations
ANOVA
df
Not surprisingly, the range for the 95% confidence interval is much larger (since we have fewer observations), but the center of the range is more or less the same. What we do note, though, is that the ranges in all three
regressions overlap quite a bit. Hence it is difficult to say that the beta has been drifting.
Regression
Residual
Total
It is also common for betas measured using daily data to be lower (particularly for less liquid stocks) because the closing price for the stock may be from a time a few minutes before the actual close, at which time the S&P
would probably be trading and for which time we'd be measuring the S&P close. So this introduces a bit of noise.
Using monthly data, I got for the period 4th November 2011 to 4th January 2016:
Regression Statistics
Intercept
SP_Ret
SS
MS
F
1 0.076964 0.076964 10.10971
51 0.388259 0.007613
52 0.465224
Coefficients
Standard
Error t Stat P-value
0.011985 -0.70049 0.486802
1.830913 0.575835 3.179576 0.002508
-0.0084
0.161516
0.026087
0.006991
Significance
F
0.002508
0.195379
53
df
Upper
95%
-0.03246 0.015666
0.674875 2.986951
Lower 95%
Significance
F
SS
MS
F
1 0.052148 0.052148 1.3661 0.247917
51 1.946812 0.038173
52
1.99896
Standard
Upper
95%
Coefficients Error t Stat P-value Lower 95%
-0.0173 0.028323 -0.61074 0.544082 -0.07416 0.039563
1.019712 0.872441 1.168803 0.247917 -0.73179 2.77121The Groupon company is discussed in the NY Times of Nov. 23, 2010. (http://boss.blogs.nytimes.com/2010/11/23/doing-the-math-on-a-groupon-deal/?
ref=business). You can also read up on Groupon's history on Wikipedia. The About Tech website has a readable description of Groupon's business model.
Groupon financial information can be found on finance.yahoo.com.
We know that a stock's beta measures its sensitivity to market movements. We also know that the stock beta is related to the nature of the product sold by a
company, especially its price elasticity.
What is your best estimate of the beta of Groupon? You can use published estimates, as long as you can explain how those published estimates have been
computed. Keep in mind, as well, that the beta is a forward-looking number. Beta estimates using historical information are backward-looking and may or may
not be the best estimate of the underlying beta, which measures the future tendency of an asset to move with the market. (As mutual funds note, "Past
performance is no indication of future results.")
Before you answer the question, read the Fortune article about Groupon's business model given below. Once you have read the article, think about Groupon's
beta in two different ways:
1. A beta measures an asset's comovement with the market. If so, we need to ask whether Groupon's profits and market value would move together with the
market or against it? Would it move in the same direction as the market, but more than the market? Or perhaps less than the market? Think carefully
about the implications of Groupon's business model for your structural estimate of Groupon's beta.
2. If Groupon's future business model is likely to be similar to its past business model, then we can use regression using historical stock returns to estimate
Groupon's beta. Download stock returns for Groupon and the S&P and estimate a regression beta for Groupon.
Based on your answers to these questions, what is your considered estimate of Groupon's beta. Remember that you can have not ony a point estimate, but also a
likely range within which you believe the true beta is likely to lie.Recall that the beta estimates the tendency of a stock to move with the market. Hence we have to ask ourselves what we think the nature of Groupon's business is - is it cyclical or countercyclical? When would a company
want to avail itself of Groupon's services? Would it be when the economy is doing well or when it's doing badly? My guess is that it would be when the economy is doing badly. If that is correct, then Groupon's sales would
be high when the market is doing badly and low when the market is doing well. This suggests a low beta or maybe even a negative beta.
As far as Groupon's costs are concerned (office costs, internet costs, personnel costs), they are probably going to move in sync with the economy. Hence that will reinforce the low beta. Although Groupon may be a "tech"
firm, does it make sense to classify it with tech firms like Google or Microsoft? Are the products that all the tech firms sell the same?
Here's what one of the students in a previous term, Philipp Zahn, said -- right on the ball, it seems to me: "The article (the 2010 NY Times article referenced above) states that many believe, that only "desperate" businesses
cooperate with Groupon. If this is really the case, Groupon might be able to find more partners in a bad economic situation."
In bad economic times, consumers are more motivated to look for goods and services that can be purchased below market price. Therefore, more customers might be willing to purchase coupons at Groupon. Groupon
would profit from a general bad economy, and return on assets would change in opposition to the companies of the benchmark. This would also cause the beta to be negative. However, consumers spend less in bad
economic times. It is uncertain, if this decrease in spending has a stronger effect on Groupon, than the fact, that consumers are more interested in low price goods and services in bad economic times.
All of this is from a purely theoretical point of view. However, since that article was published, Groupon has gone public. Hence there is now data on Groupon returns. It is, therefore, possible to see which effect is
stronger. According to Yahoo, Groupon's beta is 0.72 (http://finance.yahoo.com/q/ks?s=GRPN+Key+Statistics)
However, according to Nasdaq (May 2, 2014), Groupon's beta is 1.76 (http://www.nasdaq.com/symbol/grpn); Google declines to provide a beta. Unfortunately, therefore, we need to run a regression, ourselves, to see what
the actual measured beta is. It is possible that the measured beta is not stable, especially since Groupon has only been trading for about 2.5 years.
I regressed daily Groupon returns (capital gains plus dividends, 16th March 2015 to 15th March 2016) on daily S&P 500 returns (again capital gains plus dividends) and found the following. Note that in order to compute
returns taking into account dividends, there is a shortcut provided by Yahoo. The Yahoo site computes an adjusted closing price which includes the impact of dividends and stock splits. Hence, in order to compute the
return including dividends and stock splits, all you need to do is to simply look at the percent change in the adjusted closing price:
Regression Statistics
Multiple R
R Square
Adjusted R
Square
Standard
Error
Observations
ANOVA
Regression
Residual
Total
Intercept
SP_Ret
0.343974
0.118318
0.114806
4.443285
253
df
SS
MS
F
1 665.0005 665.0005 33.68322
251 4955.438 19.74278
252 5620.439
Significance
F
1.95E-08
Standard
Upper
95%
Coefficients Error t Stat P-value Lower 95%
-0.15815 0.279347 -0.56614 0.571802
-0.70831 0.392013
1.540367 0.26541 5.803725 1.95E-08 1.017652 2.063082ANOVA
Regression
Residual
Total
Intercept
X Variable 1
Multiple R
R Square
Adjusted R
Square
Standard
Error
Observations
ANOVA
df
Not surprisingly, the range for the 95% confidence interval is much larger (since we have fewer observations), but the center of the range is more or less the same. What we do note, though, is that the ranges in all three
regressions overlap quite a bit. Hence it is difficult to say that the beta has been drifting.
Regression
Residual
Total
It is also common for betas measured using daily data to be lower (particularly for less liquid stocks) because the closing price for the stock may be from a time a few minutes before the actual close, at which time the S&P
would probably be trading and for which time we'd be measuring the S&P close. So this introduces a bit of noise.
Using monthly data, I got for the period 4th November 2011 to 4th January 2016:
Regression Statistics
Intercept
SP_Ret
SS
MS
F
1 0.076964 0.076964 10.10971
51 0.388259 0.007613
52 0.465224
Coefficients
Standard
Error t Stat P-value
0.011985 -0.70049 0.486802
1.830913 0.575835 3.179576 0.002508
-0.0084
0.161516
0.026087
0.006991
Significance
F
0.002508
0.195379
53
df
Upper
95%
-0.03246 0.015666
0.674875 2.986951
Lower 95%
Significance
F
SS
MS
F
1 0.052148 0.052148 1.3661 0.247917
51 1.946812 0.038173
52
1.99896
Standard
Upper
95%
Coefficients Error t Stat P-value Lower 95%
-0.0173 0.028323 -0.61074 0.544082 -0.07416 0.039563
1.019712 0.872441 1.168803 0.247917 -0.73179 2.77121This estimate is even lower, but the range is even greater (with fewer observations), but again a big overlap with the other beta estimate ranges.
So the conclusion from the historical regression estimates is that the beta (for the past) is around 1 to 1.5, with a fair degree of stability. However, we'd still have to figure out if the future is likely to be similar to the past.
If we believe the Groupon business is going to remain "techy" in the future, we'd stick with this "high" beta estimate. On the other hand, if we believe the Groupon coupon business is likely to be countercyclical in the
future, we would go with a lower beta estimate.
Ultimately, we see that the final estimate still depends on our evaluation of Groupon's business. Data helps us come up with estimates, but in the final analysis, we have to apply economic analysis and our own common
sense.The Groupon company is discussed in the NY Times of Nov. 23, 2010. (http://boss.blogs.nytimes.com/2010/11/23/doing-the-math-on-a-groupon-deal/?
ref=business). You can also read up on Groupon's history on Wikipedia. The About Tech website has a readable description of Groupon's business model.
Groupon financial information can be found on finance.yahoo.com.
We know that a stock's beta measures its sensitivity to market movements. We also know that the stock beta is related to the nature of the product sold by a
company, especially its price elasticity.
What is your best estimate of the beta of Groupon? You can use published estimates, as long as you can explain how those published estimates have been
computed. Keep in mind, as well, that the beta is a forward-looking number. Beta estimates using historical information are backward-looking and may or may
not be the best estimate of the underlying beta, which measures the future tendency of an asset to move with the market. (As mutual funds note, "Past
performance is no indication of future results.")
Before you answer the question, read the Fortune article about Groupon's business model given below. Once you have read the article, think about Groupon's
beta in two different ways:
1. A beta measures an asset's comovement with the market. If so, we need to ask whether Groupon's profits and market value would move together with the
market or against it? Would it move in the same direction as the market, but more than the market? Or perhaps less than the market? Think carefully
about the implications of Groupon's business model for your structural estimate of Groupon's beta.
2. If Groupon's future business model is likely to be similar to its past business model, then we can use regression using historical stock returns to estimate
Groupon's beta. Download stock returns for Groupon and the S&P and estimate a regression beta for Groupon.
Based on your answers to these questions, what is your considered estimate of Groupon's beta. Remember that you can have not ony a point estimate, but also a
likely range within which you believe the true beta is likely to lie.Recall that the beta estimates the tendency of a stock to move with the market. Hence we have to ask ourselves what we think the nature of Groupon's business is - is it cyclical or countercyclical? When would a company
want to avail itself of Groupon's services? Would it be when the economy is doing well or when it's doing badly? My guess is that it would be when the economy is doing badly. If that is correct, then Groupon's sales would
be high when the market is doing badly and low when the market is doing well. This suggests a low beta or maybe even a negative beta.
As far as Groupon's costs are concerned (office costs, internet costs, personnel costs), they are probably going to move in sync with the economy. Hence that will reinforce the low beta. Although Groupon may be a "tech"
firm, does it make sense to classify it with tech firms like Google or Microsoft? Are the products that all the tech firms sell the same?
Here's what one of the students in a previous term, Philipp Zahn, said -- right on the ball, it seems to me: "The article (the 2010 NY Times article referenced above) states that many believe, that only "desperate" businesses
cooperate with Groupon. If this is really the case, Groupon might be able to find more partners in a bad economic situation."
In bad economic times, consumers are more motivated to look for goods and services that can be purchased below market price. Therefore, more customers might be willing to purchase coupons at Groupon. Groupon
would profit from a general bad economy, and return on assets would change in opposition to the companies of the benchmark. This would also cause the beta to be negative. However, consumers spend less in bad
economic times. It is uncertain, if this decrease in spending has a stronger effect on Groupon, than the fact, that consumers are more interested in low price goods and services in bad economic times.
All of this is from a purely theoretical point of view. However, since that article was published, Groupon has gone public. Hence there is now data on Groupon returns. It is, therefore, possible to see which effect is
stronger. According to Yahoo, Groupon's beta is 0.72 (http://finance.yahoo.com/q/ks?s=GRPN+Key+Statistics)
However, according to Nasdaq (May 2, 2014), Groupon's beta is 1.76 (http://www.nasdaq.com/symbol/grpn); Google declines to provide a beta. Unfortunately, therefore, we need to run a regression, ourselves, to see what
the actual measured beta is. It is possible that the measured beta is not stable, especially since Groupon has only been trading for about 2.5 years.
I regressed daily Groupon returns (capital gains plus dividends, 16th March 2015 to 15th March 2016) on daily S&P 500 returns (again capital gains plus dividends) and found the following. Note that in order to compute
returns taking into account dividends, there is a shortcut provided by Yahoo. The Yahoo site computes an adjusted closing price which includes the impact of dividends and stock splits. Hence, in order to compute the
return including dividends and stock splits, all you need to do is to simply look at the percent change in the adjusted closing price:
Regression Statistics
Multiple R
R Square
Adjusted R
Square
Standard
Error
Observations
ANOVA
Regression
Residual
Total
Intercept
SP_Ret
0.343974
0.118318
0.114806
4.443285
253
df
SS
MS
F
1 665.0005 665.0005 33.68322
251 4955.438 19.74278
252 5620.439
Significance
F
1.95E-08
Standard
Upper
95%
Coefficients Error t Stat P-value Lower 95%
-0.15815 0.279347 -0.56614 0.571802
-0.70831 0.392013
1.540367 0.26541 5.803725 1.95E-08 1.017652 2.063082I could simply have computed the ratio of the Covariance between the returns on Groupon and the S&P to the variance of S&P returns and I would have got the same answer much more easily. But if I did that, I would not
know how much reliance to place on my computed estimate. With a regression, I can say (assuming that the returns are normally distributed), that the true underlying beta is between 1.018 and 2.063 with a probability of
95%. This gives me quite a bit of certainty that the true beta is greater than one!
Now I went ahead and regressed weekly Groupon returns on S&P weekly returns from Nov. 7, 2011 to April 18, 2016 and found the following:
Regression Statistics
Multiple R
R Square
Adjusted R
Square
Standard
Error
Observations
ANOVA
Regression
Residual
Total
Intercept
Return
Multiple R
0.363181
0.131901
R Square
Adjusted R
Square
Standard
Error
0.128143
Observations
0.089594
233
df
Regression Statistics
SS
MS
F
1 0.281739 0.281739 35.0986
1.854255 0.008027
2.135994
231
232
This suggests a higher beta; on the other hand, this is for the period Nov. 4, 2011 to April 18, 2016. Perhaps the beta has been decreasing over time, so that a beta computed using only the last year would have a lower beta.
So I repeated the regression using the same time period as the daily regression (of course now I would have much fewer data observations). My results were as follows:
Standard
Coefficients Error t Stat P-value Lower 95%
-0.00734 0.005921 -1.23929 0.216495
1.935643
-0.019
0.326724 5.924407 1.13E-08
1.291904
0.406737
0.165435
0.149071
Significance
F
1.13E-08
0.087252
53
Upper
95%
0.004328
2.579382The CPI Card Group Inc. made an IPO filing on August 7, 2015 for $100m. Here is a description of the company from Bloomberg Business:
CPI Card Group, Inc. engages in designing, producing, personalizing, packaging, and tracking of smart cards. It focuses on providing financial, commercial, and
identification card production and related services. The company offers contactless smart card using RFID, RFID stickers, physical access cards, and FOBS,
EMV cards, and MicroSD; financial, commercial, and blank cards; microprocessor cards; and memory cards that are used for pre-paid telephone, library, access
control, parking, and loyalty programs applications. It also provides packaging customization options that include silver foil, iridescent glitter, canvas, hot foil
stamp, dimensional FX, metallic flake, and pearl. In addition, the company offers personalization services, such as daily processing for secure and non-secure
card programs; bulk DOD ink-jet printing for card programs; and B2B and B2C card issuance, distribution, and smart card and EMV personalization and
fulfillment services. Further, the company provides mailing and fulfillment solutions, such as envelope with card and carriers, greeting card with card and
greeting card envelope, self mailers, iron cross mailers, and custom designed mailers; printing of corresponding mailing materials, manufacturing and perso of
the cards, inline match verification, NCOA data processing, and custom programming; and daily fulfillment, weekly fulfillment, Web based ordering and
fulfillment for B2B and B2C, and large direct mail. It serves banking, gift and retail, prepaid, government and security, gaming and entertainment, and loyalty
and membership industries in the United States and internationally. The company was incorporated in 1982 and is based in Littleton, Colorado. It has locations
in the United States, United Kingdom, and Canada.
We know that a stock's beta measures its sensitivity to market movements. We also know that the stock beta is related to the nature of the product sold by a
company, especially its price elasticity. Keep in mind, furthermore, that a beta is not just a computed quantity, it's a characteristic of a financial asset, which can
be estimated in many ways -- the estimate using stock prices and regression being only one of them. Otherwise, the CAPM could only be used for traded
securities, which would make it a lot less useful!
What is your best estimate of the beta of CPI Card Group? You can use published estimates, as long as you can explain how those published estimates have been
computed. Keep in mind, as well, that the beta is a forward-looking number. Beta estimates using historical information are backward-looking and may or may
not be the best estimate of the underlying beta, which measures the future tendency of an asset to move with the market. (As mutual funds note, "Past
performance is no indication of future results.")
Write up your answer in a Word file using no more than two pages and email it to me by the due date. Make sure to send an Excel file; however, the Word file
should be self-explanatory. I should only need to look at the Excel file for clarification of your computations.
The rubric that I will be using for evaluation is as follows:
Rubric
Understanding of what beta is
Conceptual determination of beta by looking at product
Regression beta computation
Discussion of t-statistic/confidence interval
Use of competitors to determine CPI Card beta
Conceptual determination of beta by looking at strategy
Additional discussion/ bonus points
Total
Pts
Assigned
Max
Points
2
2
2
7
2
2
2
1
Purchase answer to see full
attachment