Dealing with the transition to high-cost (hopefully clean)

electricityV1

July 2015

Overview

• The problem: Managing the electricity shortfall• The bigger context• Short term costs and benefits• Why do we end up with second best?

The problem

• Available electricity climbed 53% to 2008, then fell by 5% as plants aged Delayed repairs - breakdowns cost 7%

of output in 2007, rising to 15% in 2014

• Response: Rationing of mining value chain, which

increased reserve margin but limited growth at height of commodity boom

Loadshedding from late 2014 Price increases Some recovery in private sector share

from around 4% in 2007-2010 to 5% today (down from 7% in 1994)-

50,000

100,000

150,000

200,000

250,000

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

2012

2014

GWh

Electricity available, year to May

Eskom Other

Changes in prices and use

Note: (a) Mostly smelters plus Sasol and paper production. Source: Calculated from Eskom, Historical average price increases. Excel spreadsheet. Downloaded from www.eskom.co.za in February 2015. Price increase calculated as average price (revenue divided by use), deflated using CPI

Municipalities, 90,000

Municipalities, 91,000

Eskom residential, 10,000

Eskom residential, 11,000

Other Eskom direct, 30,000

Other Eskom direct, 30,000

Mining, 32,000 Mining, 31,000

Eskom industrial (a), 62,000

Eskom industrial (a), 55,000

-

25,000

50,000

75,000

100,000

125,000

150,000

175,000

200,000

225,000

2008 2014

GW

H

Electricity use by type of customer

0%

20%

40%

60%

80%

100%

120%

140%

160%

Eskom residential

Eskom industrial (a)

Municipalities

Mining

Other Eskom

direct

Percentage price increase, 2008 to 2014

-

20

40

60

80

100

120

140

160

180

200

-

250

500

750

1,000

1,250

1,500

1,750

2,000

2,250

2,500

1996

1998

2000

2002

2005

2007

2009

2011

2013

thousands of GW

H

Rest of the economy

Value added, constant Rmn GWH used

-

20

40

60

80

100

120

-

50

100

150

200

250

300

350

1996

1998

2000

2002

2005

2007

2009

2011

2013

billi

ons o

f con

stan

t (20

10) r

and

Mining and smelters (b)

Value added, constant Rmn GWH used

Electricity use and value added

Context: Structural changeElectricity and growth in SA:Cheap coal-based energy was a major comparative advantage for mining and smelting• Especially from 1980s to

early 2000s due overinvestment in generation

• Eskom encouraged investment in smelters as highly energy intensive

• Special pricing arrangements to attract aluminium smelters that use imported ores

“MEC” based on close links between mining and state

Changing growth model in ‘00s:• Commodity boom encouraged investment in

mining and refining• The state:

• Focus on rolling out services to underserved (the post-colonial norm)

• Complex relationship with mining as lead sector (whither the “MEC”?)

• Buy in to “independent regulator” model without much privatisation

• Eskom needing to increase prices to cover investment costs• Push back from NERSA in name of efficiency

and restructuring• Result: Delayed investment but no change in

plans• Result: Rationing

• End of the commodity boom from 2011…

The end of the commodity boom

• The electricity crisis comes as the commodity boom ends• From late 2013 to early

2015, South Africa’s four major commodity exports saw a 40% price fall• Constrains both public and

private financing for big electricity, as well as demand -

20 40 60 80

100 120 140 160 180 200 220 240 260 280 300 320 340 360 380

1994199519961997199819992000200120022003200420052006200720082009201020112012201320142015

Apr

il 20

15 =

100

iron ore coal gold platinum

Price index, April 2015=100

Short-term choices

• The current trajectory: load shedding at least until Medupi comes fully on line• Actual load shedding

depends mostly on Extent of unplanned

repairs required Financing for diesel Maintenance of existing

co-gen contracts

OCGT OCGT

OCGT

OCGT

new supply and demand aggregation new supply

and demand aggregation

new supply and demand aggregation

MedupiMedupi

load shedding

load shedding

load shedding

Other demand

measures

Other demand

measures

0200400600800

1000120014001600180020002200240026002800300032003400360038004000

Feb-March 2015

March-July 2015

July-December 2015

2016

MW

Cost to the economyOption Amount Cost per kWh

Scheduled load shedding Up to 4000 MW R9 to R15

Diesel 2000 MW (Eskom)300 MW JHB and TshwaneEmbedded – n.a.

R4 to R5

Buy back from smelters 920 MW Asking R1,25

Co-gen 340 MW Asking R0,9

Daylight savings Up to 200 MW Small

Surcharge on higher households users

Up to 300 MW Small (only pay if do not cut use)

Eskom average wholesale R0,67

Rationing vs price increase?• Municipalities redistribute to other producers

and households– 40% of electricity use Slightly over half to enterprise, rest to households Virtually all manufacturing, retail and services

outside the mining value chain Affected by load shedding as well as disruptions

due to poor maintenance in municipalities Account for around 95% of total employment and

90% of GDP Serve around two thirds of all households

•Mines, smelters, Sasol and paper producers - 40% of electricity use Not load shed – instead held to

10% below 2008 baseline in usage since 2008

Not affected by load shedding Smelters employed around 125

000 (of which 10 000 in aluminium), with sales of R430 billion in 2014

Mines employed around 500 000 people with sales of R380 billion

• EIUG says they would prefer rationing to higher prices

• In effect: We subsidise them to stay in business?

• 5% of Eskom output direct to 4,7 million households – mostly Soweto and former so-called “homelands”

• Load shed• The remaining 15% of Eskom’s output went to other

direct buyers 50 000 commercial and 84 000 agricultural enterprises Regional users, above all the Mozal aluminium smelter in

Mozambique.

Employment by location

• Majority of employment located in metros, with Over half of

manufacturing jobs outside of smelting

Around two thirds of business services and logistics

• All municipal customers were subject to load shedding and rising prices, but not rationed -

200

400

600

800

1,000

1,200

1,400

1,600

Metros Other urban Rural

thou

sand

s

Retail trade Social services Construction

-

200

400

600

800

1,000

1,200

1,400

1,600

Metros Other urban Rural

thou

sand

s

Business services Manufacturing Transport and communications

Source: QLFS, September 2014

Employment by industry and area, September 2014

The broad options• Raise prices with mitigation through UIF

Costs: • Energy intensive users would see substantially higher costs –

around 5% increase if electricity is 20% of costs• Political pushback based on perception that Eskom is inefficient

and corrupt Benefits:• For most producers and households, electricity cost is around

3% of total – would gain more from UIF holiday than lose on electricity

• Move toward real price

• Continue with rationing Costs:• Most expensive to the

economy overall• Damaging to investor

sentiment and skills Benefits:• May be better for

refineries and some mines• Eskom and Government

do not need to find funds – fully externalised (until elections)

• Managing rationing: Could work to tailor

effects Close down aluminium

smelters? Metros playing a central

role in reducing impact

• Government subsidies Costs: • If 15% increase would be R20 bn, then would need e.g. 2,5%

increase in income tax revenues and 0,8% cut in other spending

• Continue with prices below full cost Benefits: • Cost to citizens is not transparent

What we’re doing…

• Combination of rationing (rationing of energy intensive users, load shedding for everyone else) and subsidies Lower cost for everyone But highest subsidy for aluminium Greatest benefits to mining and smelters in general

• Clearly less efficient in long run• Question is: Why?

A political economyNERSA• Act assumed privatisation – mandate is to hold down costs for

private monopolies• Still hanker for profound restructuring but no political will• Starving Eskom aims to promote efficiency, but at what cost?

The state• Gap between

government and business – how to support business AND bring about fundamental changes in ownership, employment and pay?

• Democracy with profound inequalities and divisions

• Tradition of coalition building rather than technocractic rigour - populism

Eskom• “Culture eats strategy for breakfast”• Unclear mandates and supervision• Need micro changes but no one will take it on – government also

running scared

Business• Lack of trust toward Eskom and government, especially as no

visible action to improve efficiency at Eskom• Smelters now prefer subsidy with rationing to higher price

Some big questions

• Where should the economy go in light of the end of the commodity boom and higher-cost electricity?• Prospects for mining and refining

What do we mean by “beneficiation” and what do we want from it?

Aluminium smelters?

• Is NERSA really necessary? What are the alternatives?• How to get Eskom to improve on efficiency – and

make that visible to stakeholders?

Re a leboha