Sitting in a Pretoria cafe, the lights went out and it was pitch black. But my policymaker coffee date and I just kept talking like nothing had happened.

The dangerous thing about load-shedding is that it has become entirely normal. Sure, the economy loses about R1bn a day during stage 4 load-shedding (Nova Economics estimates for Eskom) but that is far lower than estimates done in 2007 or 2019 — as the economy has adapted and the unpredictability of Eskom’s ancient fleet has become predictable.

The investment that has been dissuaded hasn’t happened and the economy has found a new (rather unsatisfactory) equilibrium. Of course, the media and political outrage gets repeated as usual. But we have yet to see meaningful shifts to actual action. Yet I have my doubts that, despite so many advisers pointing out issues for so long (including the Eskom CEO with his list of asks last week),  the problems in the system are really, fully understood by those who need to take action — politically especially.

There is no innovative or decisive policymaking. There seems a kind of happy daze that IRP2019 is being implemented (tick), energy procurement is done (bid window 5, tick; bid window 6, tick; a signing ceremony coming up for the risk mitigation round, tick).

Actually, there is serious procurement and market failure occurring that is not sinking in and so load-shedding risks being extended into the future. A significant minority of bid window 5 projects cannot reach financial close because they do not meet local content requirements as local producers don’t have the capacity and imported capital goods prices have shot up by about 30% even before the onset of the Russian invasion of Ukraine (meaning the projects now have negative returns).

There is a choice to have failure of a large number of projects in the round or continually delay it, hoping magically that capital goods prices will fall back and local content will emerge (neither will).

Bid window 6’s request for qualification and proposals then came out too early before the mess of bid window 5 was cleaned up and with many of the same potential flaws (especially in local content) in it. This is now causing confusion. The risk mitigation round similarly is underwater.

The president’s implicit admission in his state of the nation address was a positive in February that the expensive (and not to mention smell-test-failing) Karpowership deals were a non-starter when he mentioned that only 800MW out of a total of 2GW was ready to close.

Yet financial close has been delayed further waiting for the Karpowership deals to come back from the dead (they won’t, they will be bogged down in legal action forever).

This prevents other power coming on grid faster. If there were predictable, periodic procurement rounds every three to four months, then the failure of any set of projects at any time would not matter.

Instead, we are chasing our tail trying to ensure no-one fails while then locking in future procurement to the same mistakes. Equally, the lack of movement by Eskom on smaller procurement opportunities (due to a lack of action by the department of mineral resources & energy and the Treasury) or the inability for any player to push forward with fast, radically different procurement methods like feed-in tariffs all means we are locked into the current situation.

The Renewable Independent Power Producer Programme (REIPPP) does not need to be scrapped — it just needs to be reformed; the independent power producers office needs to loosen its grip on its baby. It should be made ship shape for a future where there is competition, with corporate energy buying and an independent transmission system and market operator creating a market.

There is a much deeper problem, and that is that energy procurement and the just energy transition more broadly are bound up within a political and bureaucratic mindset that all of SA’s problems must be solved through this one process — employment creation, community ownership, woman and disabled access to projects, social stability, local content and industrialisation.

The point here is not to throw all this out of the window completely — REIPPP has been highly successful at starting to answer some of these questions. The point instead is that each is not served by such a fixation that nothing ends up happening and we end up with more unemployment, projects with community ownership stuck on this merry-go-round, the demand required to create sustainable localisation not being realised.

This is why the intervention of the RMB CEO last week was so interesting — as was the polarised backlash against it. The point in scrapping local content requirements in the short term is not that localisation isn’t a fine long-term goal — it’s that you can’t create localisation in a short period of time without the demand in the first place and without the scale and predictability of rounds moving forward at pace.

For me, scrapping the targets now as well as allowing some preferred bidder projects to fail actually increases the chances of successful and sustainable localisation in the long run. Originally, the vaguely bullish view was that new procurement now coming on stream after a lag meant you just had to wait out a period of two years or so of load-shedding to get to the new nirvana.

Yet the mess procurement is in now means we are moving our own expectations of when there is new energy on grid from current rounds of procurement and so load-shedding risk starts to fall, from H2 2023 out into mid-2024. This should be deeply worrying for policymakers, with the risk of prolonged load-shedding over those crucial elections now much higher.

Many years ago I said (in these pages) that only a full blackout would galvanise the political economy to make the right decisions on energy fast enough to liberate growth and a recovery. I still believe that — as painful an outcome in the short term as it would be.

• Attard Montalto is head of Capital Markets Research at Intellidex, an SA research-led consulting company. This article first appeared in Business Day.