You are on page 1of 25

31C01300

Energy & Environmental Economics


Lecture 2: Electricity markets

Iivo Vehviläinen
Spring 2024
Why markets

• Electricity markets are an experiment based on free market


ideology in the 1970s:
– Competing private firms more efficient than public utilities

• Markets set up for the generation and use of electricity

• Transmission and distribution networks remain regulated

• Additional responsibilities for system operators to ensure that


the lights stay on

See Wilson (2002) and Cramton (2017).

1
Market model has taken over globally

• Electricity markets opened in several regions since the 1980s:


starting from Norway, U.K., U.S., and South America
• Market institutions being set up in localities across the globe:
– European Union has adopted a target of a single internal
market for electricity following the Nordic model
– Increasingly taken in to use in the developing world from
Turkey to India to Philippines
→ To understand the preconditions for a global energy transition,
a need to understand markets

2
Ideal model

“ A well-functioning and competitive power mar-


ket produces electricity at the lowest possible
price for every hour of the day…


The price formation process is therefore eco-
nomically effective for society.

→ Current market institutions are set up with the ideal market


model in mind: competition improves economic efficiency and
market allocation transfers those benefits to consumers.

Quote: Nord Pool webpages.

3
Reminder: A market serves two key functions

1. Allocate goods in the short-term


2. Market prices signal the value of the goods
– Incentives to invest
– Incentives to develop new technologies

We start with going through the allocation in the electricity market

4
Example: Demand

bid.id date.time type P Q

1 2015-01-15 11:00:00 D 0.011 144.215


2 2015-01-15 11:00:00 D 0.029 79.928
3 2015-01-15 11:00:00 D 0.042 63.523
… … … … …
79 2015-01-15 11:00:00 D 25.000 0.035
80 2015-01-15 11:00:00 D 25.010 0.464
81 2015-01-15 11:00:00 D 25.145 0.881
… … … … …
165 2015-01-15 11:00:00 D 120.900 30.000
166 2015-01-15 11:00:00 D 123.203 25.400
167 2015-01-15 11:00:00 D 126.257 45.000
at the price
of 126 someonw is
Table 1: Demand bids in the Nordic electricity market.
willin to buy
45MW 5
Example: Supply

bid.id date.time type P Q

1 2015-01-15 11:00:00 S 0.011 146.371


2 2015-01-15 11:00:00 S 0.029 272.917
3 2015-01-15 11:00:00 S 0.042 205.597
… … … … …
116 2015-01-15 11:00:00 S 20.007 4.999
117 2015-01-15 11:00:00 S 20.100 64.486
118 2015-01-15 11:00:00 S 20.200 32.611
… … … … …
583 2015-01-15 11:00:00 S 100.100 5.107
584 2015-01-15 11:00:00 S 108.000 0.569
585 2015-01-15 11:00:00 S 110.000 4.689

Table 2: Supply bids in the Nordic electricity market.


6
Solution: Market place with uniform auction

Solving the market equilibrium


Given demand bids (pi , qi )i∈D and supply bids (pj , qj )j∈S for a
single time period t, we solve:

max ∑ pi di − ∑ pj sj
di ,sj i j

s.t. dt = ∑ di , 0 ≤ di ≤ qi , ∀i,
i

st = ∑ sj , 0 ≤ sj ≤ qj , ∀j,
j

dt − st =0.

In optimum, for some quantity Q∗ , the shadow price of the balance


constraint dt − st = 0 gives the equilibrium market price P∗
7
Market place with uniform auction

Alternative interpretation
If the market price is P∗ , we can reformulate

max ∑ pi di − ∑ pj sj
di ,sj i j

⇔ max ∑ pi di − P∗ dt + P∗ st − ∑ pj sj
di ,sj i j
∗ ∗
⇔ max ∑ pi di − ∑ P di + ∑ P sj − ∑ pj sj
di ,sj i i i j
∗ ∗
⇔ max ∑(pi − P )di + ∑(P − pj )sj
di ,sj i j

But this is equivalent to surplus maximization! The market


outcome is efficient in this precise sense.

8
Example: The Nordic market

Nord Pool: Fri 27 Apr 2018, hour 12


75 demand

50
Price EUR/MWh

supply
25

0 10 20 30 40 50 60
Quantity, GW

• Demand has been inelastic ⇒ subject to change?


• Supply has more action & is the source of pollutants
⇒ going to be our focus
9
Why is demand is inelastic?

• Why do consumers stay away from the market?


– I.e., why do we see little action in the demand curve?
• Simple explanation is fixed price contracting
– If a consumer buys with a fixed price and not the market rate,
it doesn’t matter what the market price is
• Why do consumers choose to have fixed price contracts?
– Common behavioral explanation: consumers don’t understand
the benefits etc.
• But choosing fixed price contract can be fully rational
– Getting the benefits from market prices can be costly because
of, e.g., technology commitments (heating choices, industry)
or monitoring costs
• We’ll return to these questions in later lectures

10
Supply: What drives the bids that we see in the data?

Alternative explanations:

1. Ideal model: Cost of production or

2. Something else, i.e., the cost of production and distortions

11
Supply technologies and cost structures

• Several technologies to generate electricity, yet all produce the


same homogenous good*
• Technologies have different production possibilities
– Some power plants use storable fuels as inputs, e.g., coal,
natural gas, oil, or uranium
– Production in others is dependent on weather conditions, e.g.
hydro, solar, and wind power
– Plants can adjust their production over time relatively freely
(e.g. hydro) or there can be technical constraints (e.g. nuclear)
• Total cost of production different:
– Baseload and renewables: FIXED + variable cost
– Peaking plants: fixed + VARIABLE cost

*) Some differences remain in power system level characteristics (e.g., inertia), but we abstract away from those.

12
Supply structure is determined by differences in cost structures

The two straight lines show the total cost per MW for peak and baseload
plants. This is from the first reading assignment, Green (2005), see
Fig. 1, and explanations in that article.

13
Illustration: demand and supply curves in the real world

California Nordic Spain


100 100 100
Price USD or EUR/MWh

75 75 75

50 50 50

25 25 25

0 0 0

0 10 20 30 40 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60
Quantity, GW Quantity, GW Quantity, GW

Figure 1: Demand and supply bid curves for noon 3 Apr 2017.

14
Illustration: supply in the Nordic market

Table 3: Annual supply by technology in the Nordic region

HYDRO NUCLEAR THERMAL WIND


DEN 0 0 26.1 8.7
FIN 13.6 22.1 41.2 0.9
NOR 129.4 0 -4 1.2
SWE 66.1 62.6 8.9 5.8
Total 209.1 84.7 72.2 16.6

Notes: TWh/year, average value in 2001–2017. Thermal includes


generation from condensing power plants and net trade to neighboring
regions (here including net trade between countries within the region).

15
Illustration: supply in the Nordic market

Nord Pool: Fri 29 Jan 2016, hour 17


100

Price EUR/MWh
50
type
D

−50
0 20 40 60 80
Quantity, GW

• Condensing plants have high variable costs


• Combined heat and power production shares costs with heating
• Some plants have opportunity cost in addition to variable costs
– E.g., shutting down a nuclear power; output based subsidies
– For storable production, e.g., hydro: market price in the future 16
Beyond costs: Issues with the ideal model

• Market allocation distorted away from the perfect competition


ideal for various reasons:
– Policy interventions, e.g. subsidies to renewable power
(Grubb & Newbery, 2018)
– Market power of incumbents (Fabra, 2021)
– Entry frictions, e.g. nuclear licensing
– Non-convexities from physics (Wilson, 2002)
• “Missing money” problem (Cramton & Stoft, 2006)
– Assumed that installed capacity will be allowed to fall as far as
it needs to provide suppliers with high prices — regardless of
the impact on reliability
• Incentives of the retail customers (Joskow & Tirole, 2006–7)
– Consumers do not react to real time prices → prices caps and
rationing can be optimal
– All users are connected to the same network
17
Illustration: how are the bid curves formed?

Figure 2: Supply curves from the Nordic market. Do firms shift their
supply allocation across hours to maximize profits?

Source: Lundin & Tangerås, 2020.


18
Illustration: how are the bid curves formed?

Figure 3: Supply curves from Texas. Firms differ in their ability to


construct their bidding strategies

Source: Hortaçsu, Luco, Puller & Zhu, 2019. 19


Policy considerations

Can we trust the markets?

Are market outcomes aligned with the objectives of the society?

20
Policy considerations

1. Resilience towards shocks


– Preparedness for unexpected physical phenomena
– Misuses of the market structures

2. Need to decarbonize
– Intermittent renewables challenge the current power system
– Lower average prices lead to closures of “firm” capacity
– Price signals from current market model not necessarily
sufficient to coordinate capacity transition
3. Price levels can raise questions on equity

21
Recap: Market serves two key functions

1. Allocate goods in the short-term

2. Market prices signal the value of the goods


– Incentives to invest
– Incentives to develop new technologies

22
Recap: A market does not serve the following roles

1. Make decisions
– Buyers and sellers make the decisions
– Market institutions guide participants’ actions

2. Consider the impact of market actions to third parties


– These are called externalities
– Can be negative or positive
– Main focus in environmental economics

23
Agenda

This week

• Of markets, economics, and efficiency


• Electricity markets

Next week

• Things that markets do not consider: externalities

24

You might also like