So AT&T for months has been acting as if they have no idea why their network sucks in places like New York and San Francisco. But then last week the head of AT&T Wireless finally came out and admitted that the service isn’t working right — which is, like, welcome to reality, brother. Thanks for stating the obvious. But why does it suck? I mean it’s not as if there’s some weird law of physics that prevents mobile phones from working in Manhattan and San Francisco. I mean, other carriers manage to place calls there, right?

Like most things, it’s just a question of devoting resources to it.

We’ve been bitching at these guys for a long time about this, and they always play dumb, or tell us that they did some survey and found that most people love their network, or that some analyst did a study and found the network is great, and so maybe it’s the fault of the iPhone — typical corporate bullshit.

Our gripe is very simple. We’ve brought all these new customers into AT&T, and AT&T has raked in billions of dollars in revenue from these new customers, but instead of plowing that money into building out their network, they’ve held the money back and applied it to earnings — lining their own pockets and looking after their investors instead of looking after customers.

Here are four charts that may help explain the situation.

First up, take a look at AT&T’s wireless data revenues since the end of 2007, six months after we introduced the iPhone. It’s a nice smooth upward path.

Now here’s what has happened with capital expenditure over the same time period:

Hmm. Strange, no? Now look at AT&T’s operating income:

Wow. They seem to be doing all right, don’t they? And here is net income:

One weak quarter, but mostly pretty smooth right? I mean look at the last three quarters. Like the surface of a pond on a windless day, right? Weirder still, over those three quarters, AT&T’s revenues actually grew by about $700 million. And still, net profit stayed almost dead flat.

It’s an incredible performance, but you have to ask yourself, how is that these wizards can make all these big numbers work out so perfectly — yet they can’t build a network that can reliably connect calls in New York City?

Fix the network, dudes.