A few more thoughts on the 4,000 MW coal plant in India recently approved for international aid financing, which David and Joe have noted. I think this deserves attention because it’s at the center of the biggest climate question out there: how to meet tens of thousands of megawatt hours of unmet and projected power demand in India and China without huge coal plants like this Tata Mundra “Ultra-Mega” plant. It’s not simple. But following the logic for this project involves going down a “There Is No Alternative” rabbit hole.

To people in India facing daily brown-outs or a lack of electricity altogether, it may seem like environmental organizations in the U.S. are opposing this power development from a different universe. They may be. But the financiers trying to justify this project in the public interest are themselves in their own universe of self-justifying arguments.

The main justification for international aid for this project is that “super-critical” coal-generating technology will make this plant more efficient than others in India. However, the broader situation in India’s power sector is such that nearly all of the efficiency gains at the plant are likely to be eaten up by the world-beating levels of transmission and distribution loss of the rickety Indian electricity grid. It’s a good bet that the equivalent of the output of at least one of the plant’s five 800 MW generating units will disappear before it gets to an actual electricity consumer [PDF].

Why does the Indian grid lose this much electricity? After building mega-dams and large coal-generating stations, state-level utilities have been in debt for decades because they’ve never been able to recover the costs of these projects (for myriad reasons) and invest in maintenance of the grid. In turn, because electricity isn’t reliable, industrial facilities are converting to “captive generation” that further deprives utilities of their highest-paying customers.

In the state of Gujarat, where this plant is going up, captive generation capacity has been growing three times as fast as utility-generating capacity [PDF] since the early 1990s. Not surprisingly, captive generation is incredibly dirty: oil, naphtha, coal, etc. at larger scales, and diesel generators at small scales.

This is where the rabbit hole really gets twisted. Coal is supposed to be preferable because renewables aren’t supposed to be able to generate power on a gigawatt scale. But investing in coal generation and plugging it into an unreliable grid (rather than building renewables close to consumers or fixing the grid) has the effect of — get ready for this — spurring the construction of small-scale fossil fuel generation on the other end, which is not only incredibly dirty (no “40 percent efficiency gains” here) but at a scale that likely has an even smaller cost advantage, if any, over renewables.

The cost of unreliable electricity is so great for industrial facilities that I find it hard to believe that any will give up the captive generation they’ve built in favor of the marginal improvements in power supply that would come from the additional supply from this plant, spread over five Indian states. So who exactly is going to pay for all of this electricity? Lest you think the budgetary situation of state utilities in India is not of great social or environmental import, this debt is what helps prevent the investment that would actually bring electricity to the 500 million Indians who lack it, and the financing needed to do it with clean power.

Powering India and delivering energy services to millions will be a monumental task. But the Tata Mundra plant will simply be a monument to a failed approach. The history of energy mega-projects in India has not been encouraging. Mega-coal plants masquerade as stopgap measures to fill shortfalls in meeting electricity demand while the next generation of clean power can be put in place.

But by eating up financing that could go to renewables and perpetuating a dysfunctional system, it is this very approach that is holding back a more nuanced and effective approach.