Donaldson, Lufkin & Jenrette
Jamie Kiggen (jkiggen@dlj.com) 212.892.8985
Tim Albright (talbright@dlj.com) 212.892.6801
Hilary Frisch (hfrisch@dlj.com) 212.892.4374
Sender: jkiggen@dlj.com
The Internet Observer, 12.28.98
DLJ Internet Research
Return Of The Native
The idea of return is one we keep coming back to. Summer, echoes, and
Odysseus all return, while things that don't include dead letters, lost
love, and tennis serves from our father. Other things return but shouldn't,
like poison ivy and the Brothers Gibb. The notion of return on investment is
less palpable than most of these, but it's essential for understanding how
value is created. Every company, Internet or otherwise, makes at least an
implicit trade-off between spending and earnings, and the best companies
understand to the penny how their capital is deployed. So as a sequel to our
Observer of October 19th ("River Of Light"), we want to revisit the model
being created by Amazon, as an illustration of how compelling the returns of
native Internet companies can be.
First, with all the unanalytical posturing and press puffery around Internet
stocks these days, a brief review of Amazon's talents may provide some
substantive context. We've long argued that successful online retailing
demands much more than just building a Web site and offering things for sale
(some recent failures in the music category, for instance, are cautionary
proof enough). There is an underlying organizational skill set that
separates the successful online consumer products company (like Amazon,
Yahoo!, and AOL) from the hundreds more that aspire, however futile the
ambition, to equivalent achievement. This required and rare skill set begins
with an intense focus on the consumer, and it also encompasses a range of
other abilities, including branding and marketing, content development,
technology, customer service, and order fulfillment.
We'd like to add another item to Amazon's roster of retailing skills:
financial management. This isn't simply about the company having a war chest
of cash (it does) or managing street expectations so that Amazon can
continue its course of aggressive investment (it can). The differentiating
nature of Amazon's financial aptitude lies in its focus on the long-term
dynamics of its model, and in its comprehension of these model drivers. In
short, Amazon has recognized that its long-term upside lies not in
per-product profit margins, as many bearish analysts and investors seem to
think, but in the ability to accelerate its rate of return on assets and
equity.
So, if holiday celebrations and their aftermath permit, let's take a minute
and open our finance textbooks. One maxim we remember is that return on
equity (ROE) is the ultimate long-term benchmark by which managers are (or
should be) measured, and we can disaggregate the ROE equation into two
component calculations, return on assets and asset turnover (leaving aside a
discussion of the ROE rocket fuel known a financial leverage). Now, we're
happy to muscle our way into the battle about the virtues of Amazon's margin
structure versus that of a land-based retailer like Barnes & Noble. We wield
a club almost daily that pounds home the view that servers are cheaper than
bricks, mortar and employees, and we also whack the common wisdom that gross
margin degradation is necessarily inherent in the online environment. Some
fend off these blows and live to incur an even larger opportunity cost,
while an increasing number are joining the happy shareholder ranks. But what
we believe is absolutely beyond argument is that Amazon's ultimate rate of
asset turnover will be above that of almost any retailer in existence,
approached only by wealth-creation machines like Dell. Herein lies the
beauty, and the key value driver, of Amazon's model.
Before easing into the math, we can express this notion qualitatively.
Amazon supports its 4.5 million (pre-Christmas) customers with a gentleman's
farm of servers, several warehouses, a good amount of cash and about 1600
customer-focused employees. In other words, the assets underlying Amazon's
operations are not substantial, relative (for instance) to B&N's 1000 stores
and billion dollars of inventory. And because Amazon's assets are
approaching critical mass, they should begin to grow at a much lower rate
than revenue. As Amazon adds new retailing categories, the only necessary
incremental working capital goes to support the low inventory levels and
limited warehouse space of the new category. That's called asset turnover.
That's also called scalability, and Amazon has one of the most scalable
revenue lines in the known universe.
As Amazon layers new retailing categories onto its existing asset base, it
is also leveraging its existing customer base. On a per-customer basis,
adding new categories holds the prospect of accelerating the purchase
frequency and the absolute value of purchases Amazon derives from each
customer. Remember, Amazon does not have to re-acquire the customer, it only
has to direct the incremental spending stream of that customer to the new
contiguous category. As a result, Amazon captures expanded share-of-wallet
(a term usually applied to credit card market share), at minimal additional
cost.
This focus on increased share-of-wallet can be seen in some of its
innovative programs, such as its recently launched Gift-Click and Shop the
Web services (can you say "electronic commerce portal"?). Amazon's
Gift-Click enables users to select a gift and just type in the e-mail
address of the gift recipient. Amazon will then contact that individual,
confirm the mailing address, and send a gift-wrapped package with tender
words from the giver included. The idea, of course, is to expand per-user
purchase activity as it simplifies the process of giving gifts to a wide
range of people. In other words, it will increase the number of presents
sent for any occasion, which in turn increases the revenue Amazon generates
from that customer.
O.K., time to move the discussion back to the financial statements (and
since there are lots of equations to keep track of here, we'll again offer
to send along the supporting spreadsheet to those self-punishing souls who
request it). Before we get fancy, let's walk through the basic ROE
calculation. In the September quarter, Amazon achieved an annualized asset
turnover of 1.15 times, which was 50% lower than Barnes & Noble's asset
turnover of 2.25 times. If we conservatively assume a theoretical ROA for
(temporarily) unprofitable Amazon that's equivalent to B&N's core ROA in Q3
of 3.8%, Amazon's ROE is 4.4% versus B&N's 8.6%. Nothing to get real jazzed
about.
But the gap between Amazon and Barnes & Noble reverses if we refine our
analysis to include just net operating assets (NOA), which are operating
assets less cash. Of Amazon's $619 million in total assets at the end of Q3,
$557 million was a combination of cash, marketable securities and goodwill,
leaving the company with $62 million of NOA. Plugging that Amazon NOA number
into our asset turnover equation yields an annualized turnover of 11.5
times, which then amplifies Amazon's ROE to 44%. The same arithmetic for
B&N, with only $10.6 billion in cash, gets you to an ROE of 8.8%.
Since that was such a happy outcome, let's get even bolder and try to figure
out Amazon's own steady-state ROA, or more specifically, its return on net
operating assets. In "River Of Light", we argued that a true measure of
Amazon's profitability is apparent in a per-customer P&L. Our analysis
suggested that each customer generates an operating margin of 9.8%, or just
under $17 of operating income on $170 in yearly purchases. If we use
Amazon's per-customer operating margin as a plug for its normalized RNOA,
Amazon's ROE becomes an amazing 112%.
Now it's dare-to-be-great time. Our next set of metrics mirrors the
ROE/asset turnover argument, but is a more precise measure of capital
efficiency for an Internet retailer. It's called gross margin return on
investment, or GMROI (go ahead and pronounce it "jimroy" if that makes it
more memorable). Since the real operating leverage in native Internet
business models is below the gross profit line, companies that are better at
generating incremental contribution dollars should command a premium. GMROI
calibrates which companies deserve such economic praise. The GMROI formula
is a gauge of incremental gross profit over incremental investment,
inclusive of inventory, marketing, product development and administrative
expenditures.
In the most recent quarter, as Amazon prepared to launch new categories, we
saw a minus 48% GMROI, because operating expenses increased at twice the
rate of gross profit. If we net out spending for newly launched product
categories, GMROI is 4%, which is consistent with Amazon's statement that
its core books business is approaching profitability (an event that may
happen a lot sooner than you think). What's most important about GMROI (and
ROIC for that matter) is the rate and the direction of change in this
metric. Fast forward to the year 2000 (which isn't that far away) and Amazon
begins to generate three times as much incremental gross profit. This
translates into a GMROI of 205%, a number undreamed of by any land-based
retailer.
Naturally (to the extent any of this is natural), our GMROI concept leads us
directly into the promised land of return on invested capital (ROIC), a
geography that Dell's management and investors have staked out so
successfully. ROIC measures the net operating assets employed to generate
profit and is expressed as net operating profits after tax divided by
invested capital (invested capital is interpreted as operating assets, which
includes the cash required for operations; most retailers say this is 5% of
sales). Again, as Amazon accelerates out of its development spending stage
in the year 2000, required assets remain relatively fixed while sales and
gross profits continue to scale. This drives our projected ROIC from -32% in
1999 to 21% in 2001 and 32% in 2002.
Tired of being in the bark? Then let's look at the tree again. The growth of
Amazon's core customer base increases total revenue and its addition of
retailing categories increase per- customer purchase frequency, on an
operating asset base that remains relatively stable. This results in an
increase in asset turnover, which drives up Amazon's ROE and also
accelerates Amazon's gross profit growth. As Amazon's gross profitability
inflects and operating expenses drop as a percent of revenue, operating cash
mushrooms along with ROIC. Such enormous financial power allows Amazon to
rapidly develop new retailing categories, thereby capturing more customers
and starting the whole cycle over again. This is an increasing returns
business model in action.
In the end, we are brought back to the metaphorical nuances of return to
make a concluding point. The difference between Amazon's financial model and
that of a land-based retailer is like the difference between comeback and
come back. One is triumphant, the other is plaintive; one has momentum, the
other inertia; and one looks forward, while the other is retrospective. This
distinction may be a bit subtle, but as the Amazon model demonstrates its
power over time, subtlety is not likely to be its hallmark.
And speaking of hallmark, we wish all of you a very happy holiday season,
and safe journeys home.
=================================================
The DLJ Internet Observer, a biweekly research product of the DLJ Internet
Research team, is distributed through email, First Call and fax. To be
included on the distribution list simply send an email message to
eqinfomail@dlj.com with the phrase "subscribe observer" in the body of the
text, or contact your DLJ salesperson. To remove yourself from the
subscriber list, send email to eqinfomail@dlj.com with the phrase
"unsubscribe observer" in the body of the message.
Donaldson, Lufkin & Jenrette Securities Corporation, 1998. Additional
information is available upon request. |
