Thursday 30 April 2015

A dirty business at Longannet

A Scottish political issue, for which the responsibility lies at Westminster, is electricity network transmission charges. The present arrangements mean that suppliers pay to access the grid, and consumers face fairly uniform prices (across geographical areas) for their electricity. This made perfect sense in a world in which suppliers were indifferent about where they invested their capital (output for a given investment in a coal power station is independent of where the coal plant is). We now however live in a world in which yields on renewable infrastructure at some locations (e.g. windy Scotland) are vastly superior to the yields on the same infrastructure built at other locations. Do we now want to maintain this pricing structure? The polar opposite alternative is for suppliers to face the same grid access fees at each location, and for consumers to differentially pay for their electricity at a rate commensurate with their choice to locate either close to or far away from the fixed renewable resource. Given the current infrastructure, such a change in pricing would leave the average costs for the UK consumer unchanged, but the incentives for the purchasers of energy would be vastly different (which would of course change the infrastructure and hence the average costs after some time).

Differential consumer prices would, at the margin, provide an incentive for population movement from the congested South to the depopulated North. A larger effect is likely to be on industrial users of energy in capital intensive but low labour input facilities. Data centres, super-computers, server farms, aluminium smelters etc which do not currently get built anywhere because investment is preferentially going towards less efficient locations predicated upon patterns of existing demand, could be built specifically to consume the cheap energy produced in locations with great resources but which are far from current population centres.

Despite the relevance of this issue for Scottish voters in a Westminster election, it has not featured highly in the campaign. The SNP manifesto does say "transmission arrangements should work to support, rather than undermine, production of renewable energy in the most favourable locations", but the focus of this issue in the manifesto is to "press for a change to the transmission charging system that is penalising Scottish generators and threatening the future of Longannet power station". The closure of Longannet comes with job losses - so of course it's an emotive election issue.

But Longannet is a coal fired power station (one of Europe's biggest polluters), and by the argument above, it's appropriate for them to be cited on the basis of existing demand. So is the current pricing structure appropriate here, and hence the looming closure of the site? Stations like Longannet supply baseload power to balance out the peaks and troughs of renewable supply. It makes sense, from the point of view of minimising transmission losses, that the supply of baseload power in Scotland should equal the expected demand in Scotland less the expected renewable supply. If the closure of Longannet makes Scotland's supply of baseload electricity lower than the gap between its demand and expected renewable supply, then the current pricing structure is unreasonably discriminating against generators in Scotland. But if not, then a Scottish polity that has agreed to "ambitious commitments to carbon reduction" should be willing to let Longannet go, and focus purely upon gaining "transmission arrangements [that] work to support, rather than undermine, production of renewable energy in the most favourable locations".

End April Links

# Krugman discusses Brad DeLong's additional criteria, other than public good provision, for government expenditure in The hyperbolic case for bigger government. The collective outcome of individual long term choices is not what we might choose, so it's better for public institutions to be designed democratically to offset some of the flaws we make when making intertemporal decisions if the time period these are over is "long". This sounds reasonable to me. I would add that similar reasoning applies to the spatial dimension and there is a rationale for government due to the fact that it may be that democratic allocation of land resources could be more "efficient" (in the sense of maximising ex-post satisfaction with the outcomes) than decentralised private choices. Krugman describes the desirability of public schooling provision by claiming that we might under allocate resources to schooling because we overly discount the benefits that may be received in the very long term. Another rationale for public provision is that left to individual priorities, sufficient schooling may not be provided in my area because of the choices of others. These spatial externalities are a reason why goods like schools cannot be treated like tins of beans and left to the private sector, despite the fact that they are certainly rival (a place taken at a school is a place not available to another) and excludable (you don't have the automatic ability to gain access to schooling just because a school is there if they don't let you in), and so do not fit the definition of public goods.

Germany is imperiling the world economy by refusing to accept free money: "fixing the problem requires absolutely zero sacrifice on Germany's part. What the world needs from Berlin is for Germany to buy itself a bunch of nice shiny new transportation and energy infrastructure, or else for Germany to give itself a huge tax cut. Not only would shiny new projects and lower taxes be fun, but the message of the negative interest rates story is that by borrowing more money today Germany will improve its long-term fiscal situation." Not sure it's logically certain that it would improve Germany's long term fiscal position - but any positive return project makes economic sense if you can borrow at negative rates, and it's certainly possible that long term fiscal position is improved relative to the case with continued depressed demand and negative rates, where tax revenues likely to be lower than they otherwise would be because of general depressed environment.

# John Cochrane & JW Mason are exactly right when they say that while Greece needs to default, this is (or should be) entirely unrelated to whether is stays, or should stay, in the Euro.

# Roger Farmer says There is No Evidence that the Economy is Self-Correcting, showing that the unemployment rate seems to exhibit a unit root. As Krugman and Cochrane both say, this is ridiculous (it would have to go above 100% or below 0% eventually if that was actually the case). But I think Cochrane pretty much makes the point that Farmer trying to approximate with his model: "unemployment like other stationary ratios in macro (consumption/GDP, hours/day, etc.)  have important and frequently overlooked low-frequency movements".

# I'm in the Herald: Why the consequences of inequality really are severe

# Interfluidity's Tangles of pathology proposes that Liberalism, Inequality, and the "Nonpathology" of the lowest section of society into an despised underclass, form a trilemma: a society can exhibit 2 of these but not all 3. So the Nordic economies avoid pathology, but cannot offer a large spread in economic rewards; the US does offer such unequal rewards, but its losers are treated as an underclass.

The Swiss Have Eliminated The Zero Lower Bound!