Thread:
Here is an interesting analysis on PC stocks from Credit Suisse FB - don't know if it was posted, 'apologise if so:
PC Hardware & Peripherals: Wolf Trax 6 Spring Edition
08:21am EST 3-Apr-98 Credit Suisse First Boston (Wolf, Charles)
Summary and Introduction
Our new price target forecasts, updated on the basis of the most recent
quarter results, continues to show that Compaq is the most undervalued and
Dell the most overvalued stocks in our universe. Compaq's price target of
$38 offers a potential 47% gain while Dell's price target of $47 implies a
30% loss.
Before we attach a lot of significance to these results, it is important to
note that events during the March quarter undoubtedly impacted them.
Compaq suddenly found itself with excess inventories in the dealer channel
whose value was deteriorating rapidly. The market appears to have assigned
a low probability to Compaq's success in quickly liquidating them and then
delivering the double digit margins which have been a hallmark of the
company's performance in recent years. In addition, Compaq's upcoming
challenge to merge DEC's quite disparate operations with its own has no
doubt depressed Compaq's valuation.
In stark contrast with Compaq's recent woes, Dell reported yet another
superb quarter. In addition, the company undoubtedly benefited from the
market's growing perception of the sustainability of its competitive
advantage, captured in its direct distribution model. All three of Dell's
competitors in the enterprise PC market announced plans to emulate the
direct distribution model. But these efforts have proven much harder to
implement than to announce. Even if successful, they will only partially
eliminate Dell's competitive advantage. Dell would still retain a 4% to 5%
cost advantage over IBM, HP and Compaq from selling its computers directly
to customers.
Whether such an advantage justifies the magnitude of the difference in the
market's valuations of Compaq and Dell is an open question, however.
Embedded in Dell's share price is a forecast of 42% revenue growth over the
five-year period ending in 2004. In contrast, the market is forecasting
that Compaq's revenues will grow at only one-third that rate over the same
period. From a different perspective, the market expects Dell's revenues
to reach $105 billion in 2004, up from $12 billion currently. In
comparison, the market expects Compaq's revenues to "merely" double to $52
billion over the same period. In short, the market is implicitly
forecasting dramatic shifts in the market shares of the two leading
contenders for the title. On the sidelines, HP has voiced its intention to
become the number one PC vendor by 2000. And IBM's recent price actions
signal that it's no longer content to continue to lose market share to its
major rivals.
What all of this suggests is that a major inconsistency exists between the
market's implied revenue growth forecasts and its expectations of the
continuing financial health of the PC industry. We have argued that a
rational pricing environment should persist for the foreseeable future,
although events during the March quarter have sorely tested this
hypothesis. The market's forecast of market share shifts likewise raises
serious doubts as to the long-run stability of margins. It's unlikely that
the significantly shifts the market is forecasting can occur without a
disruption to the benign pricing environment of recent years. Yet embedded
in current PC company valuations is exactly this assumption.
Price Target Forecasts
Our one-year target price forecasts, shown in Table 2, start by assuming
that the PC companies covered in this report exactly hit our forecasted
numbers over the coming year. Using the output of this forecast as the
starting values in our price target model, we then forecast the revenue
growth rate, operating margin and capital growth rate each company is
likely to realize over our five year forecast period.
The number of fingers on our hand dictated the choice of a five-year
forecast. But it also has some rational underpinnings. At least for
personal computer companies, we believe the accuracy of our forecasts
deteriorates quite rapidly. Even a forecast as short as five years
stretches the bounds of accuracy. Second, the horizon markups built into
our model, as discussed in Wolf Trax 3, implicitly capture the value
investors believe a company will create beyond the forecast horizon. In
this respect, then, we at least partially compensate for our choice of a
relatively short forecast period. Finally, as shown in Table 1, the
competitive advantage periods of the PC companies, which determine the
length of a forecast, rarely exceed five years.
Our target prices represent the sum of four components.
1. The perpetuity value of the current operations value.
This number, the first component in the target price, is generated from our
forecast of a company's financial performance over the coming year.
2. The capitalized present value of the improvements in EVA over the
forecast period.
3. The present value of the markup of the current operations value at the
horizon, explained in Wolf Trax 3, and
4. Excess cash at the start of the forecast period.
Our price target forecasts are shown for the individual companies in Tables
2A through 2G. The top panel in each table reports the starting values of
the variables in our forecast, which determines the first component in our
target price. The panel also shows our revenue growth, operating margin
and capital growth assumptions.
In the middle panel, our revenue growth forecast is converted into a NOPAT
growth forecast by multiplying our assumed operating margin times the
revenue in each year. In the bottom panel, our NOPAT forecast is converted
into a stream of EVA improvements through the following formulas.
EVAt = NOPATt - (cost of capital x ending capitalt),
D EVAt = EVAt - EVAt-1,
where t = 1.... 5
This provides the stream of EVA improvements which, when discounted and
capitalized, represents the second component of the target price.
Again, as explained in Wolf Trax 3, the current operations value in the
horizon year (equal to the sum of the ending capital and EVA in that year)
is "marked up" using the regression coefficients provided by Steven O'Byrne
of Stern Stewart, to obtain the third component of the target price.
Excess cash represents the final component of the target price. This
number is obtained by adding the free cash flow generated over the coming
year to a company's current cash level and then subtracting the cash
required at the start of the forecast period. We assume that a company has
the option to use this cash to buy back shares.
The Target Prices
The key variables, which drive the target prices, are the revenue growth
rate and operating margin. Our five-year revenue growth rate assumptions,
shown in Table 1, attempt to capture the competitive positions and
opportunities of the companies in our universe.
With the exception of Apple, we expect the personal computer stocks in our
universe to grow faster than the industry over our forecast period.
International Data Corporation (IDC) has pegged the worldwide average
annual revenue growth rate at 9.4% through 2001. With growth slowing, this
would imply revenue growth of between 8% sand 9% over our forecast period,
which ends at the end of 2003.
Our revenue growth rate forecast reflects extreme selection bias. We cover
the companies who are likely to succeed and gain share. Dell, Compaq and
Gateway have grown revenues at two to four times the market's rate in
recent years. We expect that to continue. In particular, we have assumed
that Dell will show the most robust growth-at 30% compounded annually. We
expect Compaq to grow nearly as fast-at 25% annually- although recent
events at the company position this assumption on tenuous ground. We are
more comfortable assuming Gateway can grow its revenues at 25% annually
because the company has enormous upside potential in its country store
format.
We've pegged Apple's five-year growth rate at 10%. While relatively
anemic, the 10% assumption is nevertheless a heroic one in view of the
revenue implosion the company has experienced in recent years. Our bet is
that Apple will be able to stabilize its market share in its franchise
markets- creative content and education-and then grow in line with these in
1999 and beyond. To place this growth assumption in perspective, Apple's
implied revenues of $10 billion in 2004 would still lie below the $11
billion the company reported in 1995.
Our revenue growth rate forecasts for Quantum and Western Digital are
relatively less aggressive. IDC has pegged the growth of the disk drive
industry at 16% through 2001. So our forecast that Quantum and Western
Digital should grow at 20% and 17.5% respectively assumes that two
companies gain modest market share. Over the past five years the two have
done far better. As reported in Wolf Bytes 22, together, Quantum and
Western Digital captured almost 15 points of market share in the five years
ending last year.
Growth Rates Implied by Current Stock Prices
The exercise of generating target prices from growth rates can be reverse
to determine the growth rate implied by a company's current share price.
As shown in Table 1, the market expects Apple to grow revenues at a 5%
rate, considerably below the industry's forecasted growth rate, but
significantly higher than the negative growth numbers the company has
reported in recent quarters. The market likewise has subdued growth
expectations for Compaq compared with the recent past and our assumption of
25% growth. But at 13%, the market still expects the company to grow about
50% faster than the industry.
Dell is the only stock in our universe with an implied growth rate
substantially higher than our forecast. Dell's current share price implies
a five-year revenue growth rate of 42%, above our 30% forecast and five
times the industry' forecasted growth rate. To place the market's implied
forecasts in perspective, the market expect Compaq's stand-alone revenues
before the DEC merger to reach $52 billion in 2004. In comparison, it
expects Dell to reach $105 billion over the same period, almost double
Compaq's. In short, the market is implicitly forecasting dramatic shifts
in the market shares of the leading PC companies. Shifts of this magnitude
would undoubtedly create severe disruptions in pricing and turn the
industry into a hostile investment opportunity.
As implied by their share prices, the market has quite subdued revenue
growth expectations for Quantum and Western Digital. The market expects the
two to grow revenues at 10% and 12% rates respectively, substantially below
IDC's revenue growth rate forecast for the disk drive industry.
The Operating Margin Forecast
The operating margin is the other variable that importantly influences our
target prices. In our target price forecasts, we have assumed that
operating margins hold close to the levels anticipated in 1998 and 1999.
In particular, we expect Compaq and Dell to maintain operating margins of
12.5% and 10.5% respectively over the forecast period.
Since 1992, fluctuations in the operating margins of the major PC companies
have significantly subsided. Attention has turned to more efficient
capital management as the key drive of higher returns on capital. In
addition, businesses are accelerating their efforts to increase workers
access to information through networks. As a result, they have become less
sensitive to the price of the machine itself and more concerned with a
vendor's reputation along with its ability to provide network management
tools and after-sale service and support.
This change in the focus of buyers in the large corporate market, which
represents between 35% and 40% of the total market, has translated into
significant increases in the market shares of PC vendors who most closely
matched the ideal profile-the Fab Four consisting of IBM, HP, Compaq and
Dell. The combined market share of the four companies increased from 14% in
the first quarter of 1992 to 36% in the fourth quarter of 1997. With their
growth exceeding market rates, these companies have been content to
implicitly cooperate on pricing, at least through 1997.
Events during the March quarter have raised questions whether this benign
pricing environment will continue. In February, Compaq began price
promotions to liquidate several weeks of excess inventory in the dealer
channel. In contrast with past years, IBM and HP quickly matched Compaq's
promotions. As a result, Compaq was forced to take substantial reserves to
complete the liquidation. Whether the industry will return to rational
pricing once this effort is completed is an open question. Our bet is that
it will. The price elasticity of demand between the Fab Four companies is
so high that any aggressive price actions by one are bound to be matched by
the others. Nonetheless, March quarter events represent the first break in
a long string of relatively tranquil quarters.
Our forecast of operating margins of 6.0% and 5.5% for Quantum and Western
Digital respectively reflect the longer run average margins of these
companies. They clearly are higher than the depression-type margins the
companies are currently earning. Inventory corrections have rocked the
disk drive industry periodically. But with production cutbacks, they
seldom have lasted for more than two quarters. In this context, the
current correction has lasted as long as any in the past. In our opinion,
this is due to the disruptions caused by significant changes in the market
shares of some of the first- and second-tier companies. We expect the
drive market to stabilize once a new equilibrium is established, possibly
as soon as this summer or early fall.
We have assigned a slightly higher growth rate and a slightly higher margin
to Quantum than Western Digital because of its highly successful Digital
Linear Tape data storage backup product. Although Quantum could face a
competitive threat in this market, it should be able to grow this business
as fast as its disk drive business and with margins which are significantly
higher than those on drives.
Competitive Advantage Periods
A company's competitive advantage period is defined as the period during
which it is expected to earn a return on new investments which is above its
cost of capital. The accepted view is that value creating returns on
capital invite competition which, in turn, cause the return on new capital
to eventually fall to its cost. Put differently, no competitive advantage
is sustainable forever. A competitive advantage period, derived from a
stock's current price, quantifies the market's assessment of the duration
of sustainability.
From a numerical perspective, the competitive advantage period is equal to
the number of years it takes for the discounted perpetuity value of a
company's NOPAT plus the present value of its free cash flow (along with
its excess cash) to equal its current share price. The competitive
advantage periods of the companies under coverage are shown in Table 1.
They are calculated on the assumption that the companies grow revenues at
the rates implied by the current market prices.
The competitive advantage periods embedded in the current share prices of
the companies in our universe show wider variation than they have in past
quarters. In the past they were closely bunched around five years. Only
Dell and Gateway's CAPs continue close to this number.
On the short side, the market is attaching a meager 1.3 competitive
advantage to Apple's prospects. This is heroic, however, given the
company's lackluster performance in recent years. Compaq's implied
competitive advantage period of 3.3 years is also quite low, reflecting the
company's current problems in liquidating excess inventories. But it is
far shorter than a competitive advantage period which would be more
consistent with the outstanding financial performance of Compaq over the
longer run. Since 1992, for example, the stock is up fifteen fold.
The competitive advantage periods of Quantum and Western Digital-the two
disk drive stocks in our universe-are actually longer than we would have
expected. That's because the disk drive industry has been characterized by
unexpected downdrafts in margins and earnings. As a result, the visibility
of earnings in this industry has hardly extended beyond one's nose. |
