How Much Reliability Should Society Purchase?

Learning objectives

By the end of this lesson you should be able to:

Explain why perfect reliability is impossible.
Understand the trade-off between reliability and cost.
Recognise why electricity reliability is an economic as well as an engineering decision.
Understand the concepts of the Value of Lost Load (VoLL) and Loss of Load Expectation (LOLE).
Appreciate why different consumers value reliability differently.
Recognise why electricity markets increasingly seek to procure reliability rather than simply energy.

Introduction

Imagine asking a simple question:

How reliable should an electricity system be?

The obvious answer might seem to be:

"Perfectly reliable."

Unfortunately, perfect reliability is impossible.

Every additional improvement in reliability requires additional investment.

Building more power stations costs money.

Building stronger networks costs money.

Maintaining reserve generators costs money.

Keeping additional engineers on standby costs money.

Society therefore faces an unavoidable question:

How much reliability is worth paying for?

This question sits at the intersection of engineering, economics and public policy.

Reliability is not free

Imagine a town with a peak electricity demand of 100 MW.

One option is to build exactly 100 MW of generation.

This would be relatively inexpensive.

However, if one generator failed, consumers would immediately lose supply.

Alternatively, the town could build:

120 MW.
150 MW.
200 MW.

Each additional generator makes the system more reliable.

Each also increases costs.

There is therefore no single "correct" level of reliability.

Instead, society must decide how much additional resilience is worth purchasing.

The reliability-cost trade-off

In general, reliability follows a pattern like this:

Initial improvements in reliability are relatively inexpensive.
Further improvements become progressively more expensive.
Approaching perfect reliability becomes extraordinarily costly.

Imagine improving a system from:

95% reliability

99% reliability.

This may require modest investment.

Improving from:

99.99%

99.9999%

may require enormous additional expenditure for only a very small reduction in outages.

This phenomenon is known as diminishing returns.

Why zero risk is impossible

Every engineering system faces uncertainty.

Electricity systems experience:

Equipment failures.
Extreme weather.
Human error.
Cyber attacks.
Fuel supply disruptions.
Unexpected demand.
Forecast errors.

No amount of investment can eliminate every possible risk.

Instead, engineers seek to reduce risk to an acceptable level.

The objective is therefore:

Extremely reliable electricity, not perfect electricity.

The Value of Lost Load (VoLL)

One way economists approach this question is through the Value of Lost Load (VoLL).

VoLL estimates the economic cost of consumers losing electricity.

For example:

If losing one megawatt-hour of electricity causes society to lose £20,000 of economic value, then:

VoLL = £20,000/MWh.

This value includes:

Lost industrial production.
Business disruption.
Household inconvenience.
Healthcare impacts.
Economic productivity.

VoLL is not the price consumers pay for electricity.

Instead, it estimates the cost of electricity not being available.

Different consumers value reliability differently

Not every consumer experiences interruptions in the same way.

Consider three examples.

Household

A short interruption may be inconvenient.

Lights go out.

Television switches off.

Food remains refrigerated.

The economic impact is relatively modest.

Data centre

Even a few seconds without electricity may:

Interrupt internet services.
Corrupt data.
Cause financial losses.
Affect millions of users.

The economic consequences can be enormous.

Hospital

Loss of electricity may threaten life.

Hospitals therefore maintain:

Backup generators.
Battery systems.
Uninterruptible power supplies.

These consumers place an exceptionally high value on reliability.

One reliability level does not suit everyone

Because consumers value reliability differently, many critical facilities purchase additional protection.

Examples include:

Hospitals.
Airports.
Data centres.
Water treatment plants.
Telecommunications infrastructure.

Rather than requiring the entire electricity system to achieve extraordinarily high reliability, these consumers often invest in their own backup resources.

This can be a much more efficient solution.

Measuring reliability

Power systems often use statistical reliability measures.

One of the most common is:

Loss of Load Expectation (LOLE)

LOLE estimates the expected amount of time during which available generation may be insufficient to meet demand.

Importantly:

LOLE does not mean consumers will definitely experience blackouts.

Instead, it estimates the probability of supply shortages under specified assumptions.

Many electricity systems define reliability standards using LOLE targets.

Expected Energy Not Served (EENS)

Another commonly used measure is:

Expected Energy Not Served (EENS).

Rather than asking:

"How often might shortages occur?"

EENS asks:

"How much electricity is expected to remain unserved?"

For example:

A system experiencing a small shortage lasting thirty minutes has a much smaller EENS than one experiencing widespread shortages lasting several hours.

Both measures help planners compare alternative investment decisions.

Reliability standards

Most countries define minimum reliability standards.

These standards reflect society's willingness to invest in maintaining reliable electricity supplies.

The exact values differ between countries.

Some systems tolerate slightly higher levels of risk in exchange for lower costs.

Others invest heavily to minimise even very unlikely outages.

Ultimately these are public policy decisions informed by engineering analysis and economic assessment.

Capacity markets

One way electricity markets seek to improve reliability is through capacity markets.

Rather than paying generators only for producing energy, capacity markets also pay certain resources simply for remaining available when needed.

The objective is to ensure sufficient dependable capacity exists before shortages occur.

Capacity markets therefore focus primarily on:

Long-term adequacy.
Investment incentives.
Security of supply.

Whether they achieve these objectives efficiently remains an active area of debate.

Reliability is more than capacity

Although capacity markets improve resource adequacy, they are only part of the solution.

As we have seen throughout this module, reliable electricity systems also require:

Flexibility.
Reserves.
Frequency response.
Inertia.
System strength.
Black start capability.
Network capacity.

Each represents a different contribution to reliability.

Consequently, modern electricity systems increasingly procure a portfolio of reliability services rather than relying upon a single mechanism.

An emerging challenge

Historically, many reliability services were supplied automatically by conventional synchronous generators.

As electricity systems become more decentralised and renewable generation increases, these services become:

Scarcer.
More valuable.
More technology-neutral.

The challenge for market designers is therefore no longer simply:

"How much energy should we purchase?"

Instead, the question becomes:

Which combination of services provides the level of reliability society wants at the lowest overall cost?

Reliability as a public good

Reliable electricity benefits everyone.

When the electricity system functions well:

Businesses remain productive.
Public services continue operating.
Communications remain available.
Healthcare functions normally.
Economic activity continues.

Because almost every part of society depends upon reliable electricity, decisions about reliability extend beyond individual consumers.

They become questions of national infrastructure policy.

A key insight

Perhaps the most important lesson of this module is:

Reliability is not a single commodity that can simply be bought and sold.

Instead, reliability emerges from the combination of many different services operating together.

Energy.

Capacity.

Availability.

Flexibility.

Reserves.

Inertia.

System strength.

Network capability.

Restoration.

Each contributes to the overall reliability experienced by consumers.

The challenge facing modern electricity markets is recognising these different contributions and providing efficient incentives for their provision.

Key takeaways

Perfect reliability is impossible and would be prohibitively expensive.
Society must balance the cost of additional reliability against the cost of interruptions.
The Value of Lost Load (VoLL) estimates the economic cost of electricity interruptions.
Different consumers value reliability differently.
Reliability standards such as LOLE and EENS help planners assess system adequacy.
Capacity markets are one mechanism used to improve long-term reliability.
Reliable electricity requires many different services beyond energy production alone.
Modern electricity markets increasingly seek to procure a portfolio of reliability services rather than simply purchasing energy.

Module summary

Throughout this module we have seen that reliable electricity systems depend upon much more than producing sufficient energy.

Reliable systems require:

Adequate capacity.
Operational security.
Flexible resources.
Operating reserves.
Frequency control.
Inertia.
System strength.
Restoration capability.

Different technologies contribute different combinations of these services.

The central lesson is therefore:

The value of an electricity resource cannot be measured solely by the amount of energy it generates.

A reliable electricity system emerges from the coordinated provision of many complementary services.

Understanding those services provides the foundation for the next stage of this course, where we explore how electricity markets attempt to recognise, coordinate and reward these different forms of system value.