The more money you make, the more you benefit from government services. It's as simple as that.
For one, there's the old standard: infrastructure. Roads and stuff like the postal service. Have a business? You extract a lot more value from the highway system than the guy that drives 10 miles to work a day.
Education: unless you don't hire anything but illiterates for manual labor, nearly every worker you have has had 100k spent on their public education over at least 13 years. Oh, and your customers are educated too, allowing them to earn enough to buy (as well as being smart enough to want, in some cases) your products.
Law enforcement: criminals aren't just disruptive to individuals but to business as well. The implied threat of punishment prevents a lot of theft, vandalism, and fraud being perpetrated on you. A rich guy is a much bigger target for crime than the average worker. A subset of law enforcement is the court system and patents/copyright laws. There is a vast infrastructure set up purely to protect your assets and provide enforcement of abstract, artificial limits that you would in no way be able to enforce yourself (not talking about music and movies; think industrial patents and trade secrets). Without the threat of a lawsuit, what stops your competitors from making exact copies of your goods without spending the R&D costs, or keeps your debtors paying you?
Military: protecting shipping lanes and, abstractly, your life and goods from the threat of invasion. Opening up foreign markets through force (happened plenty of times throughout history), keeping stability in unstable regions to protect worldwide markets, and (let's get realpolitik here) keeping the price of oil low.
As a society, we all benefit from all these services to a greater or lesser degree, but who do you think uses more of these benefits proportionately, Joe Blow getting the median $21k a year, or billionaires like Bezos, who uses the hell out of the roads and postal service, whose business was set up by highly educated individuals, whose customers were literate [Amazon did start out as a book store, after all], that is well-known around here for attempted abuse of the patent system, that generally enjoys having secure facilities, pirate-free transportation, and the cheap oil necessary for a business based on transport? The wealthy get much more out of the government in invisible and implied benefits than any welfare queen ever could.
The fact that you saw "Obama's Katrina?" on MSNBC seems a little disingenuous, although I'll give you the benefit of the doubt, maybe you just don't watch a lot of TV news. The fact that there's a question mark means it was probably a story about this very topic, with the left-leaning MSNBC trying to discredit such a claim. A little searching finds a clip from The Ed Show on MSNBC doing exactly that.
This kind of systematized academic attention to games is long overdue, specifically in the soft sciences and humanities. Video games have now become the most profitable means of entertainment, and it is kind of amazing that so little attention is paid to them in terms of serious academic study. As a literature grad, I can tell you that many of the books I've devoted serious academic effort to have print runs that would make shovelware developers for the DS laugh. Although that's kind of an apples-to-oranges comparison (for the purpose of building analytical skills, there's nothing wrong with examining a minor work, it actually is valuable to be unable to find any prior critical work to build off of), it could help raise the esteem and perceptions of relevance for the discipline outside the Academy.
The vast majority of games have a pretty shallow narrative structure, but there are still themes, relationships, ideologies, moralities, and philosophies encoded into the choices and actions we are presented with (or instructed to make) over the course of the game. And that's not even considering the larger context of cultural markers and meanings games are embedded in. As games grow in both popularity and narrative/cultural sophistication (were we presented with choices like 'kill innocents to maintain your cover with the terrorists' before this decade? Is that a function of a maturing playerbase demanding 'realism', more awareness of the importance of moral choices by developers, shifting cultural beliefs about terrorism/warfare/entertainment in the face of omnipresent concerns this decade?), there is a lot of fertile ground to be mined in artistic and cultural criticism, and a much wider scope for psychological, sociological, and educational research. Games are not innocent or 'harmless' and have the same capacity for meaning as any fairy tale, comic book, film, or nightly news report. Actions in a video game can be seen as 'natural' consequences, rather than as constrained choices in a constructed system, much like books and films and Fox News (or any other news network, though Fox are the true masters of narrative-building) reports; what happens is justified by the internal logic and prevailing ideology of the narrative - and if you think you can see through them easily and they have no impact on your ways of thinking, just look at how many people here quote Ayn Rand or Heinlein in their sigs*. When the generation that grew up with video games from birth reaches 40, I wouldn't be surprised to see ideological quotes from games there; though games are much more indirect and admittedly don't have nearly as many choice soundbites, I wonder if the fact that a player is performing the actions himself is more psychologically effective than simply being exposed to a narrative (although the narrative depth and sophistication of even the best games pales in comparison to even merely decent writers of fiction).
We give toddlers toys to build reflexes and train physical functions, but also to help build mental pathways for things such as seeing differences in colors and shapes, cause and effect, and rudimentary knowledge of currency and careers; we can see video games as toys as well in this sense, built to entertain, but also to develop and test critical thinking and reasoning skills (well, decent games do) and to allow players to take on various roles in the world (be it mayor or covert operative). In order to build more educational games, we need to NOT look at them as textbooks but more like a laboratory. Games are for doing, not reading; hypertextual footnotes to a textbook are okay. Like conventional textbooks, however, there are implicit and unstated assumptions built into the structure of games; in order to be successful as educational tools, you need to examine the entire superstructure to try and build what you want to teach into the very playing of the game. This must be how we see games as education, not merely computerized flash cards posing math questions or multiple-choice answers providing a small playtime reward. There is a game the army is developing to teach officers how to manage villages in Afghanistan; there is a village full of people you can talk to, each with a bunch of multiple-choice questions and consequences for each one (not just in the Bioware friendly/neutral/hostile tripartite fashion). The game teaches you what to ask and how to ask it; the lesson is developing diplomacy skills through a simulated interaction, not language skills by learning rote translations.
*Okay, it's a chicken-and-egg thing here, did the person's beliefs come before or after they read the book, and of course one book doesn't (hopefully) determine the course of a person's beliefs, but we've all known that kid who got into libertarianism because he read some Heinlein at a formative age, and cumulative exposure to an ideology helps cement that way of thinking into a person.
Mystics always hope that science will some day overtake them. -- Booth Tarkington