The last article ended with Malta racing against the remaining life of Marsa Power Station. The procurement of replacement generating capacity had begun in 2006. The process took almost three years and changed along the way before the contract for the new Delimara plant was signed in May 2009. By then, however, the time available to replace Malta's ageing generating plant had become dangerously short. The consequences became painfully clear in 2010, when the problem was no longer what Malta's electricity system should look like ten years ahead. It was how to keep the existing system running until the new investment arrived.

The record of those months is worth revisiting because it shows what happens when long-term planning is allowed to run out of time. During the ten months to April 2010, Malta suffered five complete electricity shutdowns. Four of the five might have been prevented had parts of the electricity supply been switched off quickly enough when generating plant failed. In other words, the immediate challenge was no longer simply whether Malta had enough generating capacity. It was whether a fault in one part of an increasingly fragile system could bring down the whole country.

That was a very different problem from the one Malta should have been dealing with years earlier. By then Marsa was living on borrowed time. Its remaining operating hours were limited under EU environmental rules and had to be conserved, which meant relying more heavily on Delimara. But the more Malta depended on Delimara, the harder it became to take its plant out of service for the maintenance it needed. The new 144MW plant that was supposed to provide the longer-term answer was itself running late. By April 2010, the project was around nine months behind its original timetable and, even if the outstanding problems were resolved, it was at the time projected that the new plant might not be available until around June 2012.

That left Malta having to get through another two years, including two summer peaks, with what it already had. The response tells us a great deal about how little room remained. Enemalta examined leasing 20MW of temporary generation for the summer. It considered buying another turbine as a bridge until the new Delimara plant was completed. There was even consideration of bringing back into service an old Marsa turbine that had been out of use since 2000. At the same time, engineers were looking at better ways of switching off parts of the supply quickly enough to prevent a local failure becoming a national blackout.

Another proposal was to keep around 26MW of generating capacity constantly available as a safety cushion. The principle is straightforward: instead of using every machine as hard as possible, some capacity is held back so that it can step in when something else fails. The problem was cost. The estimate at the time was around €17.5 million a year in additional fuel alone. That was an extraordinary price to pay for insurance, but the alternative also had a price. The calculation was simple: that cost had to be weighed against what every hour without electricity cost the Maltese economy, particularly during the summer tourist season.

I think this is where the distinction between planning and crisis management becomes important. There was nothing inherently foolish about these measures. Faced with the system as it stood in 2010, many were sensible, and some were necessary. The people dealing with the immediate problem could not make Marsa younger, recover the years already lost, or build a new power station in a few months. Their job was to reduce the risk of another national shutdown while buying enough time for the investments already under way to arrive.

The scale of the response shows how narrow the choices had become. Existing turbines and boilers had to be brought back into service or kept running, damaged equipment had to be repaired, backup arrangements strengthened, and Delimara maintained despite the pressure to keep it producing electricity. Much of this had to happen at once and with considerable urgency. Malta was no longer choosing the shape of its future electricity system; it was trying to keep the existing one stable until the permanent investment was ready. 

This is what I mean by five minutes past midnight. Midnight was not the moment the lights went out. It had passed earlier, when Malta lost the freedom to choose calmly between different long-term options and was left managing the consequences of decisions that had been delayed. By 2010, the question was no longer whether to invest early enough to avoid the problem. The country was already spending money and taking extraordinary measures to live with it.

There is also an important policy distinction here. A well-run electricity system obviously needs good engineers, reliable plant, proper maintenance and people capable of responding when something fails. No amount of planning will eliminate every fault. But operational skill cannot substitute indefinitely for decisions about what capacity will be needed, when old plant must be replaced and how much spare capacity the country requires.

The previous article showed how Enemalta's engineering culture had become exceptionally good at keeping ageing plant alive. The events of 2010 showed the limit of that achievement.

Nor is the lesson that the 144MW Delimara extension should have been ordered a few months earlier. The problem went much further back. Demand had been rising while Marsa was ageing, and the environmental restrictions that would limit its remaining life were known well before they took effect. Yet replacement capacity entered procurement only when much of the available time had already disappeared. Once that happens, every further delay carries a much greater cost because there is no longer any room behind it.

That is why I regard 2010 not principally as a story about blackouts, but as a story about planning. The immediate measures were designed to keep the lights on, and that was exactly what the circumstances required. The policy failure was that Malta had reached circumstances in which keeping the lights on had become the overriding objective.

The aircraft carrier was starting to turn. The problem was that the rocks were already very close.

David Spiteri Gingell is a Governance, Institutional, and Digital Transformation Consultant