The myth of baseload power demand
Today’s Fin has a leader arguing that we should be laying the ground for a move to nuclear power. It’s commendably realistic about the long time lags involved, and argues we should get started on preparations now. My view is that it would be better to wait and see if the US makes progress on its (currently faltering) attempts to revive the industry there. But the thing that really got me going was the repetition of the claim that alternative energy sources are problematic because they can’t meet “baseload power demand”.
I’ve said before that this claim is wrong, but I think it’s time to sharpen my position, and state two claims:
*There is no relevant sense in which baseload power demand is a meaningful concept in our current electricity supply system.
*Any electricity supply system likely to exist in the next 40 years and capable of meeting peak power demand will have no problems meeting baseload demand.
The first point may seem paradoxical, but the reasoning is quite straightforward. Our current electricity system is based primarily on coal-fired power stations which cannot be turned on and off at short notice. So, generating power during times of peak demand (daytime) entails generating power during off-peak times, even if there is no demand for that power at a price that covers average costs. That is, we have a baseload supply, which easily exceeds the demand for off-peak power at average cost, and sometimes even at fuel cost. The result, as we observe, is that off-peak power must be heavily discounted, and even so, demand is barely enough to keep the turbines turning.
To consider any meaningful notion of baseload demand, we could do a bottom-up analysis, and consider how much of electricity demand corresponds to the notion of a continuous, stable 24/7 demand. In the average household, for example, this would include the fridge and those ‘vampire’ appliances that are left on standby all the time. In addition, of course, lots of households have off-peak hot water, but this is only because of the price incentives designed to get rid of the excess baseload supply. The same points apply to offices and a most industrial uses (including some that operate at night to take advantage of cheap power, even though other costs are higher). There are only a few continuous processes like aluminium smelting that really constitute baseload demand in the strict sense. Of course, there are off-peak demands that don’t constitute baseload in the strict sense, like people watching TV at 3am, but there’s no reason to think that such demands are large.
To get a quantitative handle, we can use the following analysis: currently off peak prices are about half of daytime prices, and offpeak demand is about half of daytime demand (illustrative numbers only, will fix). If we didn’t discount offpeak electricity, it seems likely that offpeak demand would be around a quarter of daytime demand.
So, as long as 25 per cent of supply is generated by baseload suppliers like coal, oil, geothermal and nuclear, our main problem will be one of excess baseload supply, as at present. We’re unlikely to reach that point for some decades. But even then, the offpeak demand could be met by reliable sources that are independent of time of day, most obviously gas and hydro. In that case, standard principles of marginal cost pricing would suggest that there should be no off-peak discount. In such a system, the baseload sources would be used optimally, rather than generating excess low-value electricity as at present.
A baseload demand problem would only emerge in a system reliant almost entirely (more than 75 per cent) on solar electricity. And, even if such a problem emerged, it could be dealt with exactly as we deal with our current problem of excess baseload supply, by changing relative prices.
I haven’t dealt with the separate problem of supply variability from solar and wind (hint: the answer has to do with prices, as before). But, in our current circumstances, and as regards marginal increments to the system, the far bigger problem is that of supply invariability. It is a positive disadvantage for nuclear that it generates power 24 hours a day rather than solely during the daytime. Much of that power, and the fuel used to generate it, is effectively wasted.