First, let me thank Alexandru Luta for his information on Taro Aso. Following up on his lead, I found that he had proposed back in 2009, long before Fukushima, to use subsidies to make renewables 20% of Japan's electricity by 2020, and to increase solar power 20-fold by 2020 to create a Y50 trillion industry--that's 10% of current GDP. He asserted that the economies of scale would halve the cost of solar power systems within only three-to-five years!

Alex Luta wrote:

"Innocent macro question: So, if a utility retires a coal plant and replaces it with a wind farm, is that net or gross [investment] in terms of GDP accounting?"

RK:

Good question. GDP growth is the sum of an increase in the work-hours of labor, plus the net increase in capital stock, plus improvements that increase the productivity of labor and capital, i.e. more output for a given level of input. If you replace less skilled workers with more skilled workers, that shows up in enhanced labor productivity. If you replace typewriters with PCs, that shows up in improved productivity of capital. The productivity of labor and capital combined is called Total Factor Productivity (TFP) and is the key to longterm growth, particularly in mature economies. Problems in TFP growth is one of the big reasons for Japan's economic stagnation.

In Japan, what are seeing is not the replacement of coal plants by wind farms, but rather the replacement of nuclear plants by wind farms and solar cells (and coal plants). Wind farms and solar cells produce a lot less energy per yen compared to nuclear power; hence, they LOWER the productivity of capital and thus GDP.
Wind farms also cause more carbon emissions than nuclear plants (see Lancet article that I cited the other day); if these costs were incorporated into today's GDP, they would also lower the productivity of capital and thus GDP (these costs and their consequences for GDP will show up in the future, e.g. a report in today's news about how warming of the oceans is causing fish to become smaller). Other renewables cause more pollution and thus higher health costs, and that lowers GDP.

If Japan really were replacing coal plants with wind farms--which it is not--you would have to balance out the higher cost of wind farms with the "external economies," i.e. the reduced health costs and see how they net out.

In any case, the Noda administration commissioned four different institutes to estimate the impact on GDP of eliminating nuclear by 2030. BTW, in deciding in the need for electricity, the Noda administration presumed baseline real GDP growth of about 1%; in judging the revenue to be gained from hiking the consumption tax, the Noda administration projected baseline real GDP growth that was double that pace: 2%. And so it goes.
In any case, while one government-related institute reckoned the cost to be negligible, another indusry-led one said it would make 2030 GDP Y46 trillion yen lower than the baseline scenario--almost 7.5% lower. That's almost as much as all the revenue to be gained from the consumption tax hike. Its a bit less than the anticipated increase in health care costs for the elderly during that period. The difference in the four forecasts depends on estimates of how much people lower their use of electricity due to higher costs and how much it will cost to reduce carbon emissions. It's pretty risky to hope that the "negligible" impact is the correct one. The first rule of economics--there ain't no free lunch--tells me not to buy it. But industry proponents of nukes also have an interest in playing up the downside. So, while I believe the impact of a 20-year phaseout would be severe, I have no idea how to quantify that.

In response to my comment that:

"Personally, at present, I'm inclined to see nuclear as a bridging technology until renewals are ready for prime time."

Paul Midford wrote:

"Although RK
seems to be ostensibly challenging the Noda administration policy, his position is pretty much their policy: gradually phasing out nuclear power while gradually phasing in renewables over the next quarter of a century."

RK:

Hardly. First of all, I'm not talking about a quarter century; I'm talking about several decades, depending on the pace of progress in renewables. Secondly, I'm not proposing replacing nukes with renewables in the next couple decades, but gradually replacing fossil fuels with renewables as the latter become ready with the aid of sensible and limited subsidies, and using more LNG and less coal and oil in the meantime. US exports of LNG to Japan (and others) would make this much cheaper for Japan while weakening petro-dictators.
Phasing out nukes is way down the road in my desired process. Noda, by contrast, wants to replace nuclear in the near-term with the hope of renewables and the reality of more coal plants and other fossil plants.
His goal does not even hope for a reduction in the share of fossil fuels in the total electricity picture over the next few decades. Moreover, his team is cutting back on their goal for the reduction carbon emissions. I fail to see how the fight against global warming is helped by replacing a fuel with virtually no carbon emissons, with a host of fuels with higher carbon emissions (whether they be fossil fuels or renewables, according to the IAEA).

The reason for my ambivalence about nuclear over the very long term is that, while I think nuclear can be safe IF everyone followed global best practice, that's a big IF. But I don't see how we can defeat the accelerating process of global warming while abandoning nukes in the next couple decades. As far as I can tell, the menu of history is giving us a choice among less-than-desirable meals. One can pretend canned tuna is really chateaubriand, but that doesn't make it so.

Richard Katz
The Oriental Economist