The IRP revisited: Energy modelling in our new reality

March 3rd, 2014, Published in Articles: Energize


by Johan van den Berg, SAWEA

The Integrated Resource Plan is the twenty year energy blueprint that determines broadly the energy mix of the country for the long term. But does it reflect a fundamentally transformed energy reality?

Johan van den Berg

Johan van den Berg

Wind energy was procured in Round 3 of the REIPPPP process at an average of 74c/kWh, while the latest projections on new coal power from Medupi  show a levelised cost of R1,05/kWh. Other bulk technologies are similar to, or higher than, Medupi.

In this context it was unexpected when the base case modelling of the new IRP suggested a significant drop in the installation of wind power vis-à-vis other technologies. While the ultimate recommendations suggest that government continues to procure 1000 MW of wind power annually, the adverse base case led to considerable alarm among investors, both local and international.

In principle the new model is a significant improvement over the one which informed the IRP 2010 in that it is more flexible and takes account of several scenarios. It does however show a number of shortcomings:

  • It is artificial to run a base case that ostensibly is purely quantitative. In each case, certain assumptions are implicitly made that are non-quantitative. For example, the cost of electricity should a neighbouring country be invaded to get a hold of a promising gas field or hydro resource is not modelled, even though this may be the lowest cost strategy. Once this is conceded, it doesn’t make any logical sense to exclude hard policy realities like the National  Development Plan and Green Economy Accord from the base case modelling.
  • The modelling erroneously assumed that the proposed Strategic Environmental Assessment (SEA) for wind run by the Department of Environmental Affairs would outlaw wind farms outside the special “zones” where wind power will be fast-tracked – and that capacity factors for wind power would thus drop for future wind farms.  This is explicitly not the case.
  • Rather than looking at actual levelised prices achieved in the South African context, the modelling depends on international benchmarks with assumed learning rates. These can only ever approximate reality. Wind power is presently far cheaper than any other bulk option and logically we should build as much of it as we can, as quickly as we can.
  • The externalities of coal power are significant – these have been quantified by the University of Pretoria for Kusile at between R0,97/kWh and R1,88/kWh – exceeding the levelised cost of wind power. Externalities were not taken into account in the modelling, in all probability skewing the outcome to a significant degree. As opposed to this, wind power has significant positive externalities in terms of job creation, industrialisation, community development and enterprise development (much of it inherent in the REIPPPP process and thus guaranteed).

More importantly, the IRP modelling is still being done on a basis that fails to reflect a fundamentally transformed energy reality. For decades, South Africa had a vertically integrated electricity supply industry, with Eskom responsible for generation, transmission and distribution. It also implied a certain funding model: Eskom is a national utility owned by the Department of Public Enterprises as its sole shareholder. New generation assets can only be funded from retained earnings or through Eskom issuing corporate bonds. If that is insufficient, the government, as the sole shareholder, needs to inject equity or guarantee Eskom’s obligations in some other way. The impact is that the financing is always, practically, on the balance sheet of SA Incorporated. Thus, money used to buy a power station is money no longer available to build a hospital or a school. This implies very large opportunity costs. By way of comparison, the Medupi power plant will ultimately cost about 70% of what the projected cost of the mooted National Health Insurance (NHI) would be up to 2020.

Government has had to resort to guaranteeing Eskom’s obligations to the tune of approximately R385-billion.This, as a percentage of GDP, dwarfs the international banking bail-outs after the global financial crisis in all countries bar Ireland (source: Standard Bank). Cost over-runs, too, are for our collective account. Such over-runs can be significant: At Medupi the projected cost rose from R52-billion in January 2007 to an ultimate figure of perhaps R150-billion (including interest during construction and flue gas desulphurisation) – about 45% of the projected NHI costs by 2020.

By 2015 however, some 50 Independent Power Producers will be feeding electricity into the national grid. Eskom will be one electricity generator among many.  The funding model has changed: these IPP’s sign a Power Purchase Agreements with Eskom obliging them to supply electricity at an agreed price and escalation for twenty years. This price cannot change. If the project runs over budget, that is the misfortune of the shareholders of the IPP – but never can the country be asked to pay for a cost over-run caused by the IPP. Likewise, the risk of delays is on the IPP. Up to now, the great majority of the plants are project financed and all are built with private money. The opportunity cost for the country is confined to credit support by the government to Eskom.

In the modelling which underpins the new IRP, these considerations are disregarded. Moreover, the modelling primarily uses the metric of installed costs to compare different technologies. This made sense under the old regime – but not the new. When buying ice cream, we care about the cost of an ice cream, not of an ice cream shop.

In summary, the modelling behind the new IRP seems to be sophisticated and nuanced. Assumptions and methodology can be improved to restore wind to its rightful place even in the base case. We need to fine tune, then run the considerably sophisticated model again.

Contact  Johan van den Berg, SAWEA, Tel 011 214-0660,

Related Articles

  • Will changes in tariff structure threaten solar PV rollout?
  • Wild Fire Fiasco: California utility’s decision to shut off power to avoid wild fires backfires
  • Eskom rejects Nersa’s decisions regarding MYPD4 – heads to court; Nersa responds
  • US energy giant says renewables and batteries beat coal, gas and nuclear power
  • Industry collaboration determines renewal and modernisation at NETFA