How to Learn Natural Monopoly Quickly and Correctly

How do you learn Natural Monopoly quickly and correctly without assuming that a company becomes a monopoly simply because it is large? Start with the cost of serving a whole market. A natural monopoly can arise when one provider can supply the relevant demand at a lower total cost than multiple competing providers, often because infrastructure costs are very high and average cost falls over the required output range. This eduKateSG Economics Quick Notes lesson uses a fictional neighbourhood distribution network to show how costs are shared, why duplicating infrastructure may be wasteful and why price regulation is a separate question from cost efficiency.

Define: compare the total cost of one supplier serving the market with the cost of multiple suppliers serving it.
Calculate: total cost = fixed cost + variable cost; average cost = total cost ÷ output.
Check: falling average cost at a few sample points is a clue, not full proof of natural monopoly.
Understand: marginal-cost pricing and average-cost pricing can give different financial outcomes.
Evaluate: regulation, access and quality matter even when a single network has a cost advantage.

Teaching step 1 — imagine one shared network. A fictional town has 6,000 households that need a particular network service during one period. Suppose building a provider’s network costs S$12,000, and servicing each household adds a further S$2 during that same period. A single firm serving all 6,000 households has total cost S$12,000 + (6,000 × S$2) = S$24,000. Its average cost is S$24,000 ÷ 6,000 = S$4 per household. This is a deliberately simplified cost function, not a report of any Singapore utility.

Teaching step 2 — compare two duplicated networks. Now imagine two identical firms must each build its own complete network at S$12,000 fixed cost and each serves 3,000 households. Each provider spends S$12,000 + 3,000 × S$2 = S$18,000. Together they spend S$36,000 to serve the same 6,000 households, compared with S$24,000 under one provider. Duplication adds S$12,000 of total cost in this classroom model. This is the intuition behind natural monopoly: splitting demand can make each provider’s cost per customer higher when large fixed infrastructure must be duplicated.

Teaching step 3 — see why the cheapest production arrangement does not automatically give the best price. The single network has marginal operating cost S$2 per additional household in this simplified model, but its average total cost is S$4. If a regulator required a price equal to marginal cost of S$2 for all 6,000 households, revenue would be 6,000 × S$2 = S$12,000, short of the S$24,000 total cost by S$12,000. A uniform S$4 price would raise S$24,000 and cover the assumed total cost. Thus productive cost advantage and suitable pricing are separate issues. Real networks also need investment, maintenance, quality standards and a regulatory framework.

The cost table: one network or two?

Use the same fictional 6,000-household market and the same service period in every row. The exercise assumes each independent provider must pay the full S$12,000 setup cost and faces S$2 variable cost per customer. We also assume that capacity and service quality do not deteriorate as the customer base grows.

These are strong assumptions, chosen so you can see the cost mechanism. Real projects may have network-sharing arrangements, congestion, different technologies, multiple service types and maintenance costs. A sound analysis would examine those details before claiming a real market is a natural monopoly.

One firm, 6,000 households: fixed S$12,000 + variable S$12,000 = S$24,000 total; average S$4.
Firm A, 3,000 households: fixed S$12,000 + variable S$6,000 = S$18,000.
Firm B, 3,000 households: fixed S$12,000 + variable S$6,000 = S$18,000.
Two-firm total: S$18,000 + S$18,000 = S$36,000.
Two-firm average cost across all customers: S$36,000 ÷ 6,000 = S$6.
Difference: two duplicate networks cost S$12,000 more under these assumptions.

Natural monopoly is about cost, not simply one seller

A monopoly means a single effective supplier of a product without close substitutes in the relevant market, commonly protected by substantial barriers to entry. A natural monopoly adds a particular explanation: one supplier can serve the relevant market at lower total cost than dividing the same demand across multiple suppliers.

A monopoly can exist because of patents, exclusive rights, control of scarce inputs or other entry barriers without being a natural monopoly. Equally, cost conditions could favour one network even if policy choices allow competitors to sell services using shared infrastructure. The market arrangement and the cost structure are distinct concepts.

Do not conclude that a firm is a natural monopoly just because it is currently the only provider. First compare the cost of feasible single-provider and multiple-provider arrangements for the same customer demand.

Why high setup costs make scale important

Consider the cost rule S$12,000 plus S$2 per household. With 1,000 customers, total cost is S$14,000 and average cost is S$14 per household. With 3,000, average cost is (12,000 + 6,000) ÷ 3,000 = S$6. With 6,000, average cost falls to S$4.

The initial S$12,000 is spread over more households, so the setup cost per household declines as the customer base grows. The additional S$2 service cost remains unchanged per customer in this simple formula. Falling average cost therefore persists over the output range we have examined.

A real business may eventually need an additional network, suffer congestion or face rising maintenance expenses. A textbook cost function that keeps declining indefinitely is a simplification, not a universal law of network services.

1,000 customers: total S$14,000; ATC = S$14 per household.
3,000 customers: total S$18,000; ATC = S$6 per household.
6,000 customers: total S$24,000; ATC = S$4 per household.
Interpretation: the model has a substantial cost advantage from serving more customers with one network.

What does subadditivity mean?

A more precise economic test for natural monopoly compares the cost of producing the market’s total output in one firm with the sum of the costs of dividing that output between multiple firms. If one firm’s cost is lower over the relevant demand range, economists say the cost function is subadditive for that output.

Using the 6,000-household example, one network costs S$24,000. Two separate networks, each at 3,000 customers, cost S$36,000. The one-provider arrangement is cheaper by S$12,000 under the stated cost schedule. This demonstrates subadditivity for the illustrated two-way split.

For an actual market you would need to test all feasible ways of serving it and consider technology, quality and economies of scope. Two cost points are not, by themselves, complete evidence that the whole industry is naturally monopolistic.

Price at marginal cost versus price at average cost

In a basic efficiency benchmark without externalities, price equal to marginal cost helps consumers use a product until the benefit of another unit matches the incremental resource cost. But if a natural monopoly has large fixed costs, setting price equal to marginal cost may not recover total costs.

Our fictional example has marginal cost S$2 and average total cost S$4 at 6,000 customers. At a uniform S$2 tariff, revenue would be S$12,000 and the operator would face a S$12,000 shortfall relative to total cost. At S$4, revenue equals S$24,000, which covers the assumed cost including any normal returns we put into that total.

Do not claim that a S$4 tariff is automatically the most efficient price. A regulator may consider subsidies, two-part tariffs, lifeline rates, cost-based tariffs or other approaches. Demand may also vary with price, so 6,000 customers cannot be assumed unchanged in a fully realistic pricing analysis.

Why might regulators use a price cap?

Without effective competitive pressure, a single supplier may have an incentive to restrict quantity or charge a higher price than would occur under stronger competition. One response is price-cap regulation, which limits what can be charged according to a regulatory framework and can encourage firms to reduce costs.

Another approach is cost-plus or rate-of-return regulation, where a regulator permits revenue based on eligible costs and a normal return. This can help finance essential infrastructure, but may weaken incentives to control costs if all higher costs can simply be passed through.

Price caps can also be difficult. A cap set too low may make maintenance and investment hard to fund, while an overly generous cap may allow excessive returns. Regulators may need to check quality, reliability, costs and service coverage, not just the posted price.

A cheap network is not enough if service quality fails

Imagine a one-provider network serving the town at low average cost but experiencing repeated outages or slow repairs. A cost calculation alone does not show whether households are receiving good service. Reliability, safety, response times and investment matter.

If a regulated operator can cut expenditure but reduce quality, a price-only incentive could produce an undesirable outcome. That is why real infrastructure regulation may include performance requirements, service standards and oversight.

Students should keep two questions separate: ‘Which arrangement uses fewer resources to deliver the defined service?’ and ‘Which prices, standards and rules lead to a good outcome for users?’ Both matter, but they are not the same question.

Natural monopoly versus a legal monopoly

A legal monopoly can arise from a legal restriction granting exclusive rights to supply a product or service. A natural monopoly is based on cost conditions and the size of the market, rather than the mere existence of an exclusive rule.

Some real infrastructure industries have been operated under exclusive licences, while others allow competition in retail, maintenance or complementary services despite a shared physical network. The exact legal and technological arrangement varies by jurisdiction.

Do not assume every public service is a natural monopoly or that every natural monopoly must be government-owned. An economic cost condition does not uniquely determine the ownership model.

How technology can change the answer

Suppose a new technology allows smaller providers to serve customers at much lower fixed cost. It may reduce the cost disadvantage of building a second network. Similarly, infrastructure sharing can avoid duplicating some expensive facilities while allowing firms to compete in services.

Technological progress can therefore weaken an old natural-monopoly argument. Conversely, some new platforms have large setup costs and tiny marginal costs and may exhibit strong scale economies over a relevant range.

A good conclusion is conditional: given the present demand and production technology, one provider may be cheaper. That result should be revisited when costs, demand, regulation or technology change.

Four mistakes that cause confusion

Calling every large business a natural monopoly: compare total costs at the relevant market demand.
Using fixed cost alone as proof: consider capacity, variable costs, duplicate facilities and feasible alternatives.
Confusing average and marginal cost: high fixed costs can make average cost exceed marginal cost.
Assuming S$2 marginal-cost pricing pays the bills: calculate revenue and total cost.
Assuming one provider guarantees low prices: cost advantages and pricing incentives are separate.
Ignoring future technology: the cost structure can change.

A useful correction note reads: ‘I said the network can charge S$2 because serving an extra household costs S$2. Correction: with S$12,000 fixed setup costs, that uniform price would not recover the full S$24,000 cost at 6,000 households.’

A ten-minute Natural Monopoly learning routine

Minutes 1–2: state the one-provider versus multiple-provider cost test.
Minutes 3–4: calculate the total and average cost for 6,000 households.
Minutes 5–6: repeat for two providers at 3,000 customers each.
Minutes 7–8: compare revenue under S$2 and S$4 uniform prices at a fixed quantity.
Minutes 9–10: name one regulatory trade-off and test a changed example.

If you can explain why one network is cheaper without claiming that any price is automatically fair, you have understood the main economic distinction.

Quick Economics practice

1. The fictional network costs S$12,000 fixed plus S$2 per customer. What is total cost for 6,000 customers?
2. What is average cost per customer in that one-network arrangement?
3. If two networks each serve 3,000 customers and each pays its own setup cost, what is their combined total cost?
4. In this model, why can one provider be cheaper than two?
5. At a uniform price of S$2 and fixed 6,000-customer quantity, what is revenue?
6. At a price of S$4 and the same quantity, what is revenue?
7. Does declining average cost across two observed output levels prove that every real market is a natural monopoly?

Answers and why

1. S$12,000 + S$2 × 6,000 = S$24,000.
2. S$24,000 ÷ 6,000 = S$4.
3. Two × [S$12,000 + S$2 × 3,000] = 2 × S$18,000 = S$36,000.
4. It avoids duplicating a S$12,000 fixed network setup cost.
5. 6,000 × S$2 = S$12,000, short of the assumed total cost.
6. 6,000 × S$4 = S$24,000, equal to total cost in this example.
7. No. The market’s full demand, feasible cost schedules, technology and alternative arrangements must be compared.

Frequently asked questions

Is a natural monopoly automatically a government-owned organisation?

No. Natural monopoly is a description of production costs relative to market demand, not an ownership requirement.

Why can the lowest-cost production arrangement still require regulation?

One supplier can have market power and may charge prices or choose service quality differently from what would best serve customers.

Does marginal-cost pricing always cover a network’s costs?

No. When fixed costs are large and marginal costs low, a uniform marginal-cost price may create a shortfall.

Can there be competition if only one physical network exists?

Possibly. Different providers can sometimes compete over services using shared infrastructure, subject to technical, contractual and regulatory conditions.

Keep learning with eduKateSG

For longer background, explore the How X Works Hub. Reference reading: OpenStax — Regulating Natural Monopolies; OpenStax — How Monopolies Form: Barriers to Entry; OpenStax — Monopoly Key Concepts.

Quick Notes takeaway: identify total market demand → compare one provider’s cost with multiple providers’ costs → distinguish average from marginal costs → examine prices, access and service quality separately. Cost-efficient production is not the same thing as automatically fair pricing.