This article is part of our Rising Star Portfolios series.

By tech executives' lofty standards, Larry Ellison, CEO of Oracle (Nasdaq: ORCL), has built a cult of personality, not least of which for his unabashed call-outs of competitors' products (and their shortcomings), his promotion of Sun Tzu's Art of War, and his tremendously expensive sailboats. Little surprise, then, that Oracle has built one of the more dominant moats in the tech universe.

But lately, Oracle's supremacy, once taken for granted, is in question. The refrain's been somewhat uniform: "Oracle's losing market share -- to SAP (NYSE: SAP), to Microsoft's (Nasdaq: MSFT) CRM offering, to (NYSE: CRM), and cloud-based software-as-a-service offerings." And so, the market dutifully dumped shares of the enterprise database software and hardware provider on soft sales of new software licenses in its fiscal second quarter, and the stock has not quite recovered. These concerns are real, but I believe proclamations of Oracle's death are premature.

It's true that the database and enterprise resource planning (ERP) software landscape has evolved considerably in recent years, and concerns over software as a service (SaaS) platforms' emergence are legitimate, as is the possibility that Oracle's buried its head in the sand for too long. But I'd submit that the SaaS vs. on-premise software debate is not a winner-take-all proposition. Many of Oracle's addicting qualities remain, the company continues to gush cash, and its opportunities are still rife. As the shares trade at 12 times my estimate of trailing free cash flow, the market's not given full credit to that prospect, and so I'm taking a slug equal to 3.8% of my Rising Stars Portfolio's capital -- 80% in common stock, and 20% in January 2014 $40 calls.

Oracle: database crack?
In a previous life, I spent my time peddling a niche ERP software package to large companies. Our goal was to help companies use data more effectively. Responses to my inquiries were fairly uniform: "We have Oracle [or said database software vendor], and we're not switching. We can't." After a time, it became sadly apparent, at least as it related to my incentive compensation, that Oracle is a habit companies just couldn't break. Once companies got on the Oracle, they just couldn't get off. The switching costs are positively enormous.

Oracle's built a veritable franchise on data: It offers customers database, middleware, applications, and enterprise reporting software intended to make more effective use of information. Its software and application packages run the spectrum of functions -- accounting, finance, human resources, operations, and so on. These are mission-critical items for large enterprises. Executives and employees manage reporting, day-to-day operations, and other data gathering from Oracle's systems, making them hard to replace. Because implementations are extremely data- and resource-intensive, they're also particularly hard to switch. They've afforded Oracle a valuable and recurring revenue stream in software updates, services, and consulting revenue.

Today, the crux of an Oracle investment relies on past and future: the ongoing stickiness of these revenues, a possibly transformative product strategy in Exadata and Exalogic, an SaaS vs. on-premise software war where the endgame isn't so binary as the wonks would have you think, a sizable (and as yet undeployed) war chest, and a valuation that's just too compelling in light of the perceived risks.

  • On stickiness: In the stock-picking business, there's an old adage that goes, "No one ever got fired for buying IBM (NYSE: IBM)." While threats from open-source alternatives like Hadoop and increasingly sophisticated SaaS alternatives are real, I believe that understates the impact of organizational inertia, and the enormous costs of switching from Oracle's database or middleware. To adapt an old classic: No IT exec ever got fired for installing Oracle, or sticking with it once it's done. Consistent 90% or higher renewal rates tell the story. That's culminated in an unbelievable captive, recurring, and profitable revenue stream, to the tune of $23 billion annualized.

    For a bit of context, I've estimated the value of those revenue streams: I've assumed that renewal rates approximate historical norms, that Oracle passes 3% to 4% price increases, and that selling costs are nil, because customers little choice but to renew and pay for maintenance. I've also assigned a pro-rata share of general and administrative costs, or 3% of revenue, and half of annualized R&D costs, and put a 25% tax rate on it. On this basis, I peg Oracle's recurring revenue streams' worth at $90 billion to $100 billion, or $17 to $20 a share.
  • On SaaS vs. on-premise solutions: Much ado has been made about the disruptive influence of SaaS software packages, which can accomplish the same thing as Oracle's wares at substantially lower cost. To be sure, Oracle's growth prospects will be pressured by cloud-based offerings -- but the reality is a little more nuanced. There's a price where that doesn't matter.

    For one, there is still a market for on-premise enterprise software packages in large companies -- Oracle can provide highly customized offerings, security of data, and an integrated suite of offerings. Moreover, large organizations are more likely to stick with known quantities, at least for the time. It's true that SaaS offerings will capture an ever-growing slice of the small- and midsized business opportunity, but they'll lose in circumstances where customization prevails. Equally significant, a recent JPMorgan analysis surmised that the cost of an on-premise solution approximates SaaS offerings nine years in. That's a long way of saying the debate is not that cut and dried.

    The short of it is that Oracle's bread and butter isn't burnt. The large-company opportunity remains, and moreover, Oracle's forays into cloud-based offerings via Exalogic and its Fusion public cloud offering provide opportunity to pick up lost ground.
  • A different Oracle? With Oracle's acquisition of Sun, moves into other elements of the IT supply chain (hardware, applications, and enterprise resource planning software), and the introduction of Exadata and Exalogic, it stands to reason that Oracle's is becoming not just a database company but a systems company -- much more IBM than the Oracle of yore. The intention, in Exadata and Exalogic, is to provide an integrated, end-to-end hardware, software, and analytics out of the box. For customers, there are the benefits of a product that works without lengthy and unwieldy implementations, lower IT staffing costs, and a scalable deployment model. For Oracle, it could be fundamentally transformative -- making its customers even more beholden to its wares, and the switching costs even higher. The upside's admittedly hard to estimate, but it's somewhere between large and huge: Growth in Exadata systems has been near 100% for seven quarters running.
  • A sizable war chest: At last quarter's end, Oracle had just almost $30 billion in cash, and a mere $15 billion of debt. Across its history, Oracle's proved to be a very savvy acquirer. You might think of it as a cross between a private equity incubator, hitched to a Coca-Cola-type supply chain (in its ubiquitous salesforce and customer base). Oracle acquires smaller tech firms, plugs them into its infrastructure, and increases their market opportunity vis-a-via its salesforce's reach. That means two things: First, Oracle can acquire its way into the cloud; and second, we've an interesting call option on a smart capital allocator and the potential for future share repurchases or dividend increases.

    Because Oracle can repay almost all of its debt on a single year's cash flow, the company's exhibited a cost-conscious mindset toward acquisitions, and its return on acquisitions has been quite good, I've chosen to apply a 1.2 times multiple to Oracle's cash balance (after adjusting for taxes on repatriation). On this basis, I think its cash hoard is worth at least $5 per share.

Add it up
Between Oracle's cash and the value of its installed base alone, I estimate the company's worth to be about $22 to $25 a share. That means we're paying next to nothing for potential growth. More importantly, the downside risk of lost growth opportunities; customer erosion from SaaS offerings, SAP's HANA, or others; or a large botched acquisition are relatively muted at today's price. It also doesn't give account to the possibility for growth from acquisitions, cross-sells to existing customers, or a successful foray into the cloud. Given a longer-term view, I expect that Oracle can grow its revenues 6% and maintain its operating margins, and under this condition, I believe the shares are worth $40.

Short and simple: Oracle has one of the widest moats in the software biz, and Larry Ellison's one fearsome tech pirate. And that's why I'm buying.

Interested in cash cows with dominant moats? Check our free report detailing three stocks that will help you retire rich.

This article is part of our Rising Star Portfolios series, where we give some of our most promising stock analysts cold, hard cash to manage on the Fool's behalf. We'd like you to track our performance and benefit from these real-money, real-time free stock picks. See all of our Rising Star analysts (and their portfolios) here.