Seeking Alpha
Profile| Send Message|
( followers)  

I recommended Oracle (ORCL) as a strong buy on 6/11/2012 when it was trading around $27 a share. Since then, the stock price has climbed up to a high of $36.31 on 2/1/2013, close to its 3-year high of $36.5. Recent financials agree with my investment thesis about the company. I rebuilt my DCF model, which now estimates a fair value around $47, more than 30% higher than the current market price of $34.90. This stock offers sound growth at a very reasonable price and therefore remains my favorite. I'm confident that its current price offers enough margin of safety in the long run. However, one should decide whether to buy it now or wait for a better entry point between $30 and $33 in the upcoming market pullback.

Outlook on fundamentals remains strong

I maintain the same opinion about Oracle's growth prospects as I did in May 2012; those are reiterated below.

Transferable strengths will help Oracle win the cloud battle for three reasons: First, Oracle excels at database. Multi-tenancy and big data are two strengths of Oracle. Second, sticky products with high switching costs will continue to help Oracle win deals from big corporations. Lastly, Oracle really knows how to grow through acquisitions.

A minor change of my opinion is that I no longer expect any noticeable margin contraction over the next few years, as I come to realize that the challenge of cloud computing on traditional software business, or at least Oracle, has been overstated. My reasons are presented below in response to some typical bearish remarks about Oracle. Data referenced in the article, unless otherwise specified, is sourced from Morningstar.com or SEC filings.

1. "Faux" or not, Oracle is fully equipped to meet client needs on cloud computing

This article argues that Oracle offers "fake" cloud and customers will eventually abandon its products to reduce costs. However, in my opinion, cloud computing itself is an obscure and abstract technical concept. You can get this feeling by reading its definition on Wikipedia. It means a variety of things for different types of customers, such as Google Drive or iCloud for retail customers, Amazon Web Services for SMEs, and Software-as-a-service (SAAS) for large corporations. Whether Oracle's cloud offering is "faux" or not has nothing to do with the two critical questions: what Oracle's clients want and whether Oracle products are able to meet their needs.

Oracle serves large corporations who rely on its products to run critical business functions. Reliability and business continuity, security and privacy, and integration with existing infrastructures are their top concerns. Absolute cost-saving and flexibility in pricing remain secondary. Oracle's robust software and middleware product lines are very capable of addressing such primary concerns.

Reliability and business continuity. These two qualities have long been core strengths of Oracle's products. Few people, even those who are skeptical of Oracle's continued success, doubt that.

Security and privacy. Security is a heightened concern in the era of cloud computing, according to many industry surveys, such as Intel's survey about what is holding back the cloud in May 2012. Page 9 of the survey indicates that customers are less concerned with private cloud than public cloud. Public cloud is hosted in the vendor's data centers, while private cloud is hosted on servers behind customers' own firewalls for security or compliance reasons. Over the next few years, it is reasonable to expect a greater adoption of private cloud than public cloud. As reported, Oracle clearly recognizes this trend and is able to win clients, who cannot migrate to public cloud, from its competitors. Regarding public cloud, Oracle rolled out Oracle Database 12c on 9/30/2012 to support multi-tenancy and took extra care to "make sure your [clients'] data is isolated and private and safe and secure".

Integration with existing infrastructure. Oracle's products have notoriously high switching costs. Since its products are used to support critical business operations, migrating away from them requires extensive planning and testing, including even running both Oracle and new systems in parallel for an extended period of time. Such project can easily take a few years to complete. The aforementioned article quotes Cisco (CSCO) as an example of cost-saving by switching to private cloud. In my opinion, this example precisely highlights the enormous time and resources required for a switch. It took Cisco four years to complete the migration. Interestingly enough, the deployment was aided by Oracle database and middleware. According to the article, Mike Myers, Director of Technical Services for Data Centers and Platform Services at Cisco said that "the addition of Oracle database schemas and middleware to the service catalogue made a huge difference in how useful the environment could be."

2. Oracle's incumbency advantage sees no sign of deterioration.

Oracle's unmatched advantage is built on its huge client base and its wide and deep product offerings, which are expanding through R&D and acquisitions. There is no sign indicating that such economic moat is deteriorating.

Large customer base. It has 390,000 customers around the globe and it serves the entire Fortune 100 roster and top 20 firms in many industries. It makes $16.21 billion just from license update and support of its existing clients in the fiscal year ended in May 2012. This alone accounts for 3.86% of global spending on data center systems and enterprise software combined in 2012, which is $419 billion according to Gartner. Furthermore, its software updates revenue strongly correlates with the new software license revenue of the previous fiscal year, as seen below in Table 1. This indicates that Oracle continues to make a handsome amount of money out of existing clients.

Table 1. Yearly Revenue of New Software Licenses and Software Updates

Year Ended May 31(Revenues in $millions)

2006

2007

2008

2009

2010

2011

2012

New software licenses

4,897

5,874

7,501

7,112

7,525

9,220

9,235

Software updates and product support

7,027

8,541

10,507

11,997

13,175

14,876

16,210

Software updates and product support (predicted using the multiplier of 1.79, 6-year average between 2007 and 2012)

8,749

10,495

13,401

12,706

13,444

16,473

Software update revenue over previous year new software license revenue

1.7441291

1.78873

1.5993867

1.8525028

1.9768771

1.7581345

Successful acquisitions. Oracle is arguably one of the best acquirers in the industry. Its website provides a nice summary of all the acquisitions it has done since 2005. These deals are organized by product lines. This article analyzes the Eloqua (ELOQ) deal in December 2012 and demonstrates how Oracle can snatch a niche player right before it turns profitable. Another article looks at its acquisition of Acme Packet (APKT) announced recently and concludes that this deal is the final missing piece of its industry solution for communication.

Despite the economic moat mentioned above, investors are still worried that Oracle may become irrelevant all of a sudden because of cloud computing. Many industries have undergone paradigm shifts over the past few years, such as mobile, brick-and-mortar retailing, and consumer electronics. Such changes have hit some big names hard. Just to name a few: BlackBerry (BBRY), Nokia (NOK), Best Buy (BBY), Barnes & Noble (BKS), Dell (DELL) and Kodak (EK). Here are two articles back in 2011 that best summarized the anticipated threat of cloud computing to software industry, particularly enterprise software. Looking back, I believe such threat is overstated and investors' fear is unwarranted. First of all, the big three in enterprise software, Microsoft (MSFT), IBM (IBM) and Oracle have embarked on their journey to cloud. Most importantly, on the demand side, Oracle differs from those fallen ones because Oracle's customers are large corporations, who are biased towards the status quo and have to consider existing infrastructure investments, as opposed to individual customers, who can be very fickle and fashion-driven. For a case in point, switching from Nokia to iPhone is nowhere comparable to switching from Oracle Fusion to Salesforce.com.

3. Oracle's hardware problem is well-contained and can be fully accounted for in valuations

Oracle's hardware revenue dropped by 9.25% y-o-y in F2012, 19.47% in Q1 F2013 and 16.29% in Q2 F2013. Therefore, many critics claimed that Oracle's acquisition of Sun Microsystems in 2010 is a complete failure. This article covers stories from both sides in a fairly neutral tone while the other article written by Bob Evans, an Oracle employee, fervently argues against Oracle bears. In short, Oracle's explanation is that the smaller yet promising engineered systems business, which sells mainly Exa products, is growing more slowly than the bigger but doomed commodity hardware business is shrinking. Oracle's management team has foreseen this trend and, to some degree, it intends to get rid of the commoditized hardware business that has no future. Larry Ellision, CEO of Oracle, said that "selling systems loaded with Oracle intellectual property, along with deemphasizing the selling of low-margin undifferentiated products like commodity x86 servers and LSI disk storage systems, products that contain no intellectual property, those 2 things have reshaped and downsized our hardware business, while making that business much more profitable" on the Q2 2013 earnings call. On the same call, he also suggested that hardware will "turn the corner" next quarter.

The following three reasonably indisputable facts convince me that setting conservative growth estimates on hardware revenue over the next five years should be sufficient to account for this problem in valuations. Operating margin should not be negatively impacted as the high-margin engineered systems business is replacing the low-margin server business.

  • Engineered systems business is growing fast but from a smaller base. It brought in $274 million in Q4 F2012, which implies an annual run rate that is at least $1 billion now, given the 70% sequential growth on unit bookings during the most recent quarter. However, it only accounts for about one-sixth of the $6.3 billion hardware business in F2012.
  • Hardware accounts for a relative smaller portion of Oracle's topline, only 17% in F2012. A 10% decline in hardware sales translates into a 1.7% decline in total revenue. Simply speaking, its poor performance is not enough to sink the boat.
  • Table 2 shows growth figures by business segments. It is clear that the hardware problem does not have any spill-over effects on other segments. Larry's remark quoted above also implies that there are few cross-selling opportunities between Oracle's flagship software business and the low-margin hardware products that do not carry Oracle-branded software.

Table 2. Revenue Growth by Oracle's Business Segments

Revenue Growth (y-o-y)

2011-5

2012-5

2012-8

2012-11

6 month ending 2012-11

New Software License

22.59%

7.27%

6.43%

16.65%

11.76%

Software License Updates and Product Support

13.02%

9.56%

2.73%

6.87%

4.90%

Hardware

203.23%

-9.25%

-19.47%

-16.29%

-17.77%

Services

19.00%

1.21%

-5.65%

-4.75%

-5.17%

Total Revenue

32.82%

4.21%

-2.25%

3.43%

0.63%

Model Assumptions

My model in June 2012 forecasted free cash flows for each year between F2012 and F2021 and assumed a 3% perpetual yearly growth rate for the FCF after 2021. The annual revenue growth rate ranged between 9.72% and 12.49%. Operating margin was assumed to decrease from 35.28% to 25.48% by 2021. Tax rate was predicted to stay at 26.43%.

I have rebuilt my model with revised outlook on revenue growth, margins and tax rates. The new model only forecasts FCF for each of the next five years to be consistent with my models built for other companies. Below are key model assumptions. Long-term free cash flow growth is kept at 3%. Details about other model inputs and justifications about my assumptions can be found here.

In short, the base case assumes a 5-year CAGR of 4.08% and the bull case, 6.79%. The bear case assumes an annual revenue growth rate of 1.5%, half of the average annual growth rate of U.S. GDP since 1947, which is 3.25%. In the base case, I predict revenue growth for each business segment and sum these revenues up to arrive at the total revenue. I assume hardware business continue to decline until 2016. My estimates of Oracle's revenue growth rates between now and 2017 are relatively conservative, compared to Gartner's estimated growths on data center systems and enterprise software, which are around 4.5% and 6.5% over the next five years respectively.

I set the cost of equity to be 10.5% to reflect the increased market risk. WACC is then used as the discount rate. For cross-check, in the base scenario, Oracle's EBIT margin decreases from 32.85% in 2012 to 30.25% in 2017 and its net margin decreases from 26.89% to 24.66% in 2017. FCFF over net income remains very close to 131.20%, the 2012 level, and so does FCFF over total revenue.

Table 3. Key Model Assumptions for ORCL

Revenue growth

2012

2013

2014

2015

2016

2017

5-year CAGR

Base

4.21%

2.31%

2.22%

3.91%

5.46%

6.60%

4.08%

Bull

4.00%

6.00%

8.00%

8.00%

8.00%

6.79%

Bear

2.31%

1.50%

1.50%

1.50%

1.50%

1.66%

Operating Margin

2013

2014

2015

2016

2017

Base

36.92%

37.00%

36.75%

36.50%

36.25%

36.00%

Bull

37.00%

37.00%

37.00%

37.00%

37.00%

Bear

37.00%

36.50%

36.00%

35.50%

35.00%

Effective tax rate

23.00%

24.56%

24.56%

24.56%

24.56%

24.56%

EBIT margin (base)

32.85%

31.96%

31.53%

31.11%

30.68%

30.25%

Net margin (base)

26.89%

26.05%

25.71%

25.36%

25.01%

24.66%

FCFF over revenue (base)

35.28%

33.90%

33.55%

33.21%

32.87%

32.53%

FCFF over net income (base)

131.20%

130.10%

130.54%

130.99%

131.45%

131.93%

Conclusion

At the current price of $34.90, it offers an upside potential of 30% to 50% with limited downside risk. Particularly, my model incorporates very conservative estimates on revenue growth for both base and bear scenarios. Even so, the bear case suggests that ORCL is trading at a $5 discount. Alternatively, if I lower the long-term FCF growth rate to 1%, the model in the bear case scenario now estimates a fair value of $30.53 and suggests valuations ratios in 2017 that are very close to Oracle's 5-year averages. It is very unlikely for Oracle, arguably the best software company in the world, to grow at the rate of 1%-1.50% assumed in the worst case scenario. What is more, my model assigns a long-term FCF growth of 3% for Oracle and 5% for Infosys (INFY) and INFY is considered fairly valued. Oracle's fair value estimate is revised up to $70, implying a 100% upside, if its perpetual growth rate is revised up to 5%.

Nowadays, investors really hate uncertainty because one cannot easily tell the difference between a value play and a value trap; think about Corning (GLW). Investing in growth-at-reason-price stocks seems to be a strategy that offers the best of both worlds. Oracle happens to be a textbook example of GARP stocks, because it offers solid growth at a rich discount to its fair value. I'm comfortable with the margin of safety provided by the current price of $34.90. However, one could opt to wait for a better entry point between $30 and $33 as the market pullback unfolds.

Scenario

Bull case

Base case

Bear case

Worst case

Fair value estimate per share

$54.00

$46.20

$39.81

$30.53

Terminal P/E

24.97

25.15

25.34

18.10

Terminal P/CFO

18.51

18.49

18.47

13.19

Terminal P/S

6.63

6.44

6.24

4.46

Upside Potential/Downside Risk

54.74%

32.38%

14.06%

-12.53%

Long-term FCF growth

3%

3%

3%

1%

Source: Oracle Is A Decent GARP Stock That Offers 30 Percent Upside