German Wind Capacity Revisited: High Cost versus Least Cost

By Donald Hertzmark -- September 7, 2010

My post last week evaluated the claim that wind generation can save money for power pool customers.  It was found that the supposed savings could be realized only if the elephant in the room – the above-market feed-in tariff – was ignored.  In other words, consumer payments for electricity from a power pool was half of the story; the real price had to include the consumer-qua-taxpayer funding of the feed-in-tariff (FIT).

And with this two-part scheme, games are played. Wind generators can bid a low price into the pool only to receive a higher FIT, which gives them an incentive to underbid. This might reduce the pool price but not overall cost to Germans for electricity.

Investing in New Generation: What Makes Sense?

If a generation resource is a good investment for its developers then it must return a profit to them.  In a normal electricity market this profit comes from supplying a segment of the demand (peak, intermediate/cycling, baseload) from a plant that is efficient technically and financially.

For existing plants and determinations of electricity costs in the here and now we can figure out the average cost of supplying electricity by calculating the weighted average cost of supply for each time period in the market every day.  If the addition of one generation source raises this weighted average without improving service quality or reliability, then it is not economical and would generally not be chosen in a well-functioning market.

But what about the future?  Electricity suppliers must invest large sums in new generation plants with the expectation that these plants will meet demand at the least cost.  This cannot be known with certainty, and mistakes are made all the time, especially when government policy and rent-seeking drive investment choices.

Transmission network operators – those in charge of the “natural monopoly” part of the power business – try to reduce the risk attendant to future supply by figuring out the least costly way to supply power and energy to their customers in the future, including the wires to transmit the electricity.  They have to take account of a long list of considerations: investment cost, fuel supply, emissions and licensing regulation, proximity to existing load centers and transmission nodes, transmission congestion – you get the idea.

The transmission system operator also has to pay attention to public policy – renewable energy mandates (“portfolio standards”), federal tax incentives (producer tax credits for wind and solar), feed-in tariffs, powerful politicians who do not want their vistas impaired – in a host of ways that directly impact their views of an optimal future generating system.

What Does the Wise Transmission Operator Do?

A wise investor in generation will first figure out what is economic to build? what are the physical constraints on the system? and finally, what limitations will public policy put on otherwise least cost generation choices?

A Case Study of “Germania”[i]

Let us imagine that we have a rather large and wealthy country to play with, one that currently has about 129 GW of installed generation capacity.  Further, we can imagine that this wealthy country, responding to its powerful environmental movement, has decided to

(i) phase out nuclear power;

(ii) limit future coal power-plant operations;

(iii) build a lot (a lot!) of wind generation plants; and

(iv) bring in most of its gas supply from Russia at prices linked directly to refined oil products and crude (i.e., high and volatile).

Such a country would have a great deal of baseload generation capacity – coal + lignite + nuclear – perhaps half of total generation capacity (US has coal + nuclear capacity of about 40%).  Suppose further that green thinking had created incentives (FIT) that pushed wind up to about 20% of total nameplate capacity.  Most of the country’s hydro generation and imports are soaked up by wind mirroring and shadowing.

With all of that generation capacity essentially independent of fuel price trends it is no accident that the cost of generating electricity in Germania is (i) high; and (ii) not responsive to changes in oil and gas prices.  At current world oil prices the average cost of electricity generation in Germania is about 6.7¢/kWh (7.8¢/kWh if crude reaches $110/bbl).

Germania’s Least-Cost Generation System

With natural gas still expensive – kind of like burning oil but more efficiently – what does a least cost generating system for Germania look like in 2020, about the time the initial wave of early “teens” investments go on line?

The first thing you do is throw out the phase outs – keep the existing nuclear and efficient coal plants in operation until you have a more cost-effective substitute;

Second, phase out your highest cost oil plants – heavy fuel oil and old combustion turbines (2.6 GW)

Third, build some new, more efficient coal plants (1.6 GW) and CCGT units (4.5 GW), and nuclear (4 GW), reduce emissions per kWh by more than 30% in the new coal plants;

Fourth, operate existing wind (28 GW) but do not build new generation from that source.

Total annual cost in 2020: $48 billion at 6.7¢/kWh for average supply cost and 6.6¢/kWh for new supply.

Back to Our Previously Scheduled Programming

Everyone in Germany’s power sector knows that this least cost system is simply some renegade economist’s fantasy.  Back in the Real World of Energy Policy, there are several important considerations that must be accommodated:

Stay clean – phase out at least 25% of older coal and lignite plants, make permitting of new coal plants difficult;

Stay green – stay the course on wind energy; and

No nukes – continue to phase out nuclear power.  Build only enough new nukes to keep the Gauls happy.

So what does Germania get for its $53 billion annual outlay for electricity (at 7.6¢/kWh average supply cost and 7.8¢/kWh for new supply)?

Less coal and lignite – existing units fall to 45 GW from 58 GW, new coal rises to 3.1 GW to make up for some of the lost legacy capacity;

Less nuclear power – existing nuclear capacity falls to 15 GW (from 19 GW), new nuclear capacity falls to 2.7 GW from 4 GW in the least cost case;

More gas – 6.5 GW of new CCGT, up from 4.5 GW in least cost case;

More wind – another 17.5 GW, for a total of 35.5 GW of wind, 25% of total nameplate capacity; and

More imports – enlarged interconnection with Benelux and Denmark/Norway is only cost effective way to shadow additional wind and meet peak demand.

And if the price of oil rises to $110/bbl by 2020, then this system will cost Germania $62 billion annually at 8.8¢/kWh.

The True Green Scenario – phase out 50% of coal and nukes, double wind installations, increase imports – costs about 7.9¢/kWh on average, 8.3¢/kWh for new supplies and carries annual costs of $56 billion.  This option requires 15.6 GW of new CCGT, 2.1 GW of new combustion units, 10.6 GW of import capacity and operates existing HFO units at 100% of capacity (!), even building a couple of new HFO plants to meet demand – not green, not cheap and not feasible (it only solves inside the box).  And by the way, higher oil prices for this beauty will cost $66 billion/y at 9.3¢/kWh.

Does ‘Green’ Always Have to Hurt?

Easing off the green pedal a bit creates enough breathing room in Germania to generate a lot of clean electricity at a much lower cost.  A more moderate program, even with 10 GW of new wind, can be done at a far lower cost with just a few adjustments:

Slow the phase out of existing coal and nuclear plants – keep 85% of existing coal and nuclear plants in operation in 2020;

Build new coal and nuclear plants – reduce emissions per kWh and burn less coal overall, and improve the efficiency and security of the nuclear fuel cycle;

Import more – let more medium term supply come from lower cost Benelux and Scandinavian suppliers to mirror/shadow wind and follow load.

A more moderate program of this sort could supply Germania for about $49 billion annually at an average cost of 6.9¢/kWh.  Even higher oil prices do not hurt as much as in the more aggressive scenario, with $110/bbl crude oil increasing total annual costs to $56 billion at 8.0¢/kWh.[ii]

As a Great Philosopher Once Said, “A man’s got to know his limitations”

In a world of unlimited wealth, where electricity can be stored and plants can be built instantaneously on a whim, a complete remake of a large power system seems feasible and even desirable to some.  Back in the real world, where everything takes time, costs money and different sources of electric power are not perfect substitutes for one another, such ambitions are difficult to realize.

Germania set up too many targets on too short a time frame.  The result was a series of conflicting mandates and constraints – close it down; no, we need it, keep it open; will the Jutes cooperate on supply? What happens to our gas supply when “bad weather” rolls in from the East (as it eventually does in Germania)?

A cleaner, greener power supply system is possible over a longer period of time at a far lower cost than a crash program.  Orderly replacement of older coal plants with more efficient and cleaner new ones makes sense given the country’s resources, geography and demand patterns.  As in the US a crash program to overhaul the system will ultimately force increased reliance on older, less efficient coal plants; it ignores microeconomic rationality for chimerical goals and wastes a lot of money and energy in the process.  Or to paraphrase that revered Soviet philosopher, “you may not be interested in markets, but markets are interested in you.”  Germania ignores market forces at its own peril.

---

[i] The simulations in this section come from a model of least cost generation that endogenizes some categories of risk.  The model is called “Port Opt for Generation”.  This program is a medium term optimization that includes time of day demand, wind shadowing, generator characterization, imports/exports and a variety of parameters with regard to coal, nuclear, hydro and HFO use and construction/phase-out.  Different risk parameters, including oil prices, technology prices and operational characteristics can be modeled explicitly.

[ii] The cost of new supply is below the average cost of supply by about 7% for this moderate scenario, while the “Green” future shows new supply above average cost by about 2-3%.