We’ve been asked many times recently why spot prices are so high, so this has prompted me to finally put fingers to keyboard for another blog. The spot prices are a function of low hydro storage, restricted fuel supplies, and plant outages.
By Greg Sise, 24 October 2018
The following chart illustrates the distribution of monthly average spot prices at the Haywards node in Upper Hutt using data up to and including February 2018, and going right back to October 1996 when the spot market started. The average prices are shown with labels; the blue line is the minimum monthly average price for each month; and the red line is the maximum monthly average price for each month.
This “envelope” around spot prices has remained stable for almost a decade, with the overall maximums above $300/MWh being set in the 2008 dry year. Interestingly, though the highest average price is in May at Haywards, it is followed closely by February and March, a function of a significant number of dry summer-autumn periods.
Contrast this with the latest update of this chart which includes the monthly average prices up to and including 23rd October.
Even if conditions were to return to normal tomorrow, the average price for October is not going to turn out below $220/MWh, so we are looking at a new record for this time of year. Furthermore, the high prices are looking like they could continue for the next month, depending on whether we get significant inflows into the hydro lakes in the next week or three: high inflows could see prices come out much lower in November.
The first chart above uses actual market prices over the last 22 years of the spot market, and up until last year, they made September and October look relatively benign in terms of price volatility. But our forecasts use inflow data back to 1931, and for two decades they have consistently shown a number of years which also create very high prices in these months, and in fact through to December: the 1976 inflow year is a good example. So high prices at this time of year are not a complete surprise to us, since 1996 we just haven’t had an inflow event to create them.
However, the prices we’re seeing at present are not just the result of low inflows and falling hydro storage, there are other factors at play which are not only almost impossible for us to predict, but which also appear to be unprecedented in the history of the electricity spot market.
From time to time over the last 22 years the market has experienced gas outages, restricting supply of gas to gas-fired generators. But this year we are experiencing the second of two long outages and what is more, they are both the result of problems with the same gas supply: the pipeline bringing gas from the offshore Pohokura field.
Based on media reports, the field operator, Shell NZ, says the current outage is likely to continue to near the end of November. Let us hope they are right and that it does not last longer.
The Pohokura gas field is owned by Shell NZ, OMV and Todd Energy, and commenced production in 2006. It has consistently produced around 70 PJ since 2007, about two thirds of which comes from the offshore part of the field. The offshore pipeline, carrying gas to the onshore production station, had its first problem back in April when bubbles were detected leaking form the pipe. This caused an outage which lasted through to July.
The current outage started at the end of September when a valve on the offshore pipeline failed. This has restricted gas supply which appears to have limited the ability of some gas-fired generators to reach full output. Genesis Energy is burning more coal than expected, and is so concerned about the fuel situation that it is now about to import coal to keep its ageing Rankine units (steam turbine units) running while also maintaining its coal stockpile at adequate levels. This will cover for the outage of the company’s 400 MW “e3p” plant that is about to go offline for about six weeks for planned maintenance.
Contact Energy’s Stratford gas-fired peaking station had a prolonged outage which was due to continue through November, but which has ended for the time being: it looks like it will be completed next year instead.
Snow melt is now in full flow, which is helping to keep the lakes from falling faster than they would otherwise, and demand is falling as we move through spring. But inflows are running below average in both islands, so storage is now at the 15th percentile, the fourth lowest level on record for late October: from now through to mid-January is when the lakes are normally filling up, so any deficit now has to be made up over the next two to three months.
On a brighter note, assuming the Pohokura gas outage ends in November, and we get some nor’westers that bring rain to the southern hydro lakes, then the spot prices will quickly return to more normal levels.