There are two big things that can help electric cars (does Musk also include plugin hybrids?) 1) like you said, is the cost of gas. It's not going to stay at $3.50 or $4 a gallon, in 20 years, we may well have hit oil's terminal decline phase, and even if not, the cheap stuff is gone. Expect at least $10/gallon in current dollars in 20 years. 2) the cost of batteries. They're going down. Fast. see: http://www.plugincars.com/lithium-ion-battery-prices-drop-160-kwh-2025-123193.html I actually used the 8% growth in energy density claimed above, which brings my price to $220/kWh in 2025 (not $160 like mckinsey claims). At that price, a 100 mile range battery will cost $8000, down from about $20,000 today. By 2032, prices could come down to $4500 for a 100 mile range. This assumes exponential gains in capacity/density/etc, but those gains have happened in the last ~15 years. The big question is whether they'll continue in the future. I'd say we'll get to about $250/kWh before battery scientists slow down, but improvements won't stop. That will get us to some pretty cheap, reasonable range electric cars in 20 years. Basically, the Nissan Leaf will go from $35,000 ($20k for the car, $15k for the battery) to $25,000 ($20k for the car, $5k for the battery) [assumptions: no EV efficiency gains, still 35 kWh/100miles, which is what the Leaf, Volt, and plug in Prius achieve today.]

The next 1 gigawatt nuclear plant built in the west will cost 5-10 Billion dollars. Look at Finland's effort: http://www.nytimes.com/2009/05/29/business/energy-environment/29nuke.html three years or more behind schedule and easily 50% over budget. The economics and short work schedules are what make renewables more attractive than these hulking plants designed with the 1960s mindset. Smaller plants have much lower impact if they go down, and can usually get back online faster than a large coal plant (1 day) or a nuke (1-4 weeks). I used to like nukes, being a technocrat, but the economics don't work out. They actually never worked out. Over the history of the grid, total capital expenditures have been roughly equally divided between generation (power plants), transmission (high voltage, long distance lines) and distribution (the lower voltage lines on wood poles bespoiling your suburb). In fact, transmission was 10-20% more than the other two, which tells you the problem with giant wind farms in the Dakotas. The exception to this was the 1970s, when our current fleet of nukes was built, and generation took up almost 50% of that CapEx pie. So be happy that your utilities are encouraging everyone to save energy rather than build new plants, because we need end use efficiency to get us more cold beers and hot showers for less energy, before we need more power plants, nukes included.

What do you mean by taxing generally? Huckabee's huge sales tax? How about controlling externalities. I see the slippery slope of taxing sin, but isn't every tax bad and discouraging, so we might as well raise revenue in this way. Governments around the world are proposing bans on incandescent bulbs in the next 5 years. A tax is a cleaner, revenue-raising alternative. Think Amory Lovins' feebates http://en.wikipedia.org/wiki/Feebate/ if they have to be revenue neutral.