Markets do not run for free. Every exchange — buying, selling, contracting, coordinating — consumes resources beyond the nominal price: search, negotiation, enforcement, and the risk of things going wrong. Those are transaction costs. Ronald Coase asked why firms exist when markets could in theory coordinate everything: his answer was that sometimes the cost of using the market exceeds the cost of doing the work inside an organisation. When transaction costs are high, integrate. When they are low, use the market. The boundary of the firm is set by where transaction costs tip the balance.
Transaction costs come in several forms. Search and information costs: finding counterparties, learning prices, assessing quality. Bargaining costs: negotiating terms, drafting contracts. Policing and enforcement costs: monitoring performance, resolving disputes. Hold-up risk: one party invests in relationship-specific assets and the other exploits that dependency. Each of these can be large enough to kill a deal or to make internalisation cheaper than outsourcing. The practical question is always: how big are these costs here, and who bears them?
Technology and institutions change transaction costs. The internet slashed search costs for many goods. Standard contracts and reputational systems cut bargaining and enforcement costs. When transaction costs fall, more activity moves to markets — more outsourcing, more gig work, more modular supply chains. When they rise — trust collapse, legal complexity, asset specificity — activity moves back inside firms or into long-term relationships. Startups that "disintermediate" are often just reducing transaction costs in a segment where they were high.
The strategic implication: build or partner based on where transaction costs sit. If transacting with outsiders is cheap, stay lean and outsource. If it is expensive — because of specificity, information gaps, or enforcement problems — bring the activity in-house or lock in long-term contracts. The same logic applies to product design: features that reduce transaction costs for the customer (one-click buy, clear pricing, dispute resolution) can win markets.
Over time, the boundary of the firm is not fixed. As technology, regulation, or trust shifts, transaction costs rise or fall — and the optimal boundary moves. The rise of cloud computing lowered the transaction cost of buying compute and storage, so many firms stopped running their own data centres. The rise of platform marketplaces lowered the cost of finding and transacting with freelancers or sellers, so more activity moved outside the firm. Watch for secular shifts in transaction costs in your industry; they often precede big changes in who does what.
Section 2
How to See It
Transaction costs appear wherever exchange is costly beyond the sticker price. Look for friction in finding, agreeing, or enforcing deals. Look for parties who avoid the market and do things themselves or through tight relationships. Look for industries that consolidated or disaggregated when technology or regulation changed the cost of transacting. High transaction costs often explain why "obvious" market solutions do not appear.
Business
You're seeing Transaction Costs when a company brings a previously outsourced function in-house. The nominal cost of the vendor may be lower, but search costs (finding and vetting suppliers), contracting costs (negotiating SLAs), and hold-up risk (dependency on a critical partner) make the total cost of transacting higher than doing it internally. The firm's boundary shifts.
Technology
You're seeing Transaction Costs when a platform reduces friction for its users — one-click checkout, instant verification, escrow. Each reduction in search, payment, or trust cost makes more transactions happen. Amazon's buy box and Prime reduced the cost of "should I buy from this seller?" and "when will it arrive?" That is transaction cost engineering.
Investing
You're seeing Transaction Costs when an investor prefers concentrated positions in companies they know well. The cost of finding, evaluating, and monitoring many small positions can exceed the benefit of diversification. High transaction costs (information, attention) push toward concentration. Low costs (index funds, liquid markets) push toward diversification.
Markets
You're seeing Transaction Costs when a new marketplace or platform takes off. Often the insight is "transaction costs in this market were too high." Real estate, freelancing, payments — the winning entrant typically cut search, matching, or trust costs. The size of the opportunity is the size of the transaction cost that was eliminated.
Section 3
How to Use It
Decision filter
"Before making vs buying, or before designing a product or market, ask: what are the full costs of transacting? Search, negotiation, enforcement, hold-up risk. If those costs are high, integrate or lock in long-term relationships. If they are low, use the market. And ask: can we reduce transaction costs for our customers and capture value from that?"
As a founder
Make vs buy is a transaction cost decision. Outsource when the market is thick, quality is observable, and contracts are easy to enforce. In-house when the activity is specific to you, when quality is hard to verify, or when a partner could hold you up. The same applies to your product: if your users face high transaction costs (finding suppliers, trusting counterparties, paying across borders), building tools that cut those costs can be the wedge. Stripe reduced payment transaction costs. Airbnb reduced trust and search costs for lodging. Map where transaction costs are high in your space; that is often where the opportunity is.
As an investor
Companies that lower transaction costs for a large market can scale fast — the demand was there but was blocked by friction. Evaluate new businesses by asking: what transaction cost are they reducing, and how big is the cost they are removing? Also ask: does this business itself have high or low transaction costs with its own suppliers, partners, and customers? High internal transaction costs limit scalability and margins.
As a decision-maker
When considering a partnership, acquisition, or outsourcing deal, price the full transaction cost. Not just the fee — the cost of finding and vetting, negotiating, monitoring, and the risk of renegotiation or failure. If the total cost of transacting is high, consider internalising or choosing a different structure. When designing processes, ask where you are imposing transaction costs on others (approvals, forms, unclear terms) and whether you can reduce them.
Common misapplication: Assuming that the market price is the full cost. The sticker price is one component. Add search cost, negotiation cost, enforcement cost, and risk. Often the "cheaper" external option is more expensive once transaction costs are included. Run the full cost comparison.
Second misapplication: Ignoring how your strategy changes transaction costs for others. If you make it easier for customers to find, trust, or pay you, you are reducing their transaction costs — that can be a moat. If you add friction (complex pricing, poor support, lock-in), you increase their cost of leaving, but you may also increase their cost of choosing you in the first place.
Amazon systematically reduced transaction costs for buyers (search, trust, delivery) and sellers (fulfilment, payments, reach). The marketplace and Prime lowered the cost of "will I get what I ordered?" and "when?" FBA lowered the cost for third-party sellers to reach customers. The firm's boundary — what Amazon does itself vs what it enables through the platform — has been shaped by where Amazon could reduce transaction costs at scale.
Netflix reduced the transaction cost of watching a film: no trip to the store, no late fees, no decision per rental. The shift from "transaction per movie" to "subscription for access" collapsed repeated transaction costs into one relationship. The same logic applies to content licensing: Netflix internalised content creation and ownership where repeated negotiation and hold-up risk made market transactions costly.
Section 6
Visual Explanation
Transaction Costs — The full cost of an exchange = price + search + negotiation + enforcement + risk. When total cost of using the market exceeds the cost of doing the activity inside the firm, the firm's boundary expands.
Section 7
Connected Models
Transaction costs explain firm boundaries and market design. The models below either formalise the idea (Coase), describe related frictions (switching costs, information asymmetry), or apply it (make vs buy, contract design).
Reinforces
Coase Theorem
The Coase Theorem says that absent transaction costs, parties would bargain to the efficient outcome regardless of initial rights. In the real world, transaction costs are positive — so initial allocation and institutional design matter. Coase's work on the firm is the same idea: transaction costs determine whether coordination happens in the market or inside the firm.
Reinforces
Friction
Friction is the resistance to movement or change. Transaction costs are a form of friction in exchange — they slow or block deals. Reducing friction in a process (onboarding, payment, support) is often reducing transaction costs. The two concepts reinforce each other in product and operations.
Tension
Information Asymmetry
Information asymmetry raises transaction costs: the party with less information pays more to search, verify, and protect against exploitation. Reducing asymmetry (signals, warranties, transparency) lowers transaction costs. The tension: full transparency may not be feasible or desirable; the optimal level of disclosure is where marginal cost of hiding information equals marginal transaction cost saved by revealing it.
Tension
Switching Costs
Switching costs are a transaction cost of leaving one provider for another. High switching costs lock in customers but can make acquisition harder (new customers face the same cost of switching to you). The tension: building switching costs reduces your customers' transaction cost of staying but may raise their cost of choosing you in the first place if they anticipate lock-in.
Section 8
One Key Quote
"The main reason why it is profitable to establish a firm would seem to be that there is a cost of using the price mechanism. The most obvious cost of 'organising' production through the price mechanism is that of discovering what the relevant prices are."
— Ronald Coase, The Nature of the Firm (1937)
Coase identified the cost of using the market — discovering prices, negotiating, contracting — as the reason firms exist. When that cost is high, production is organised inside a firm instead of through a series of market transactions. The quote is the seed of transaction cost economics: the boundary of the firm is drawn where the cost of the next transaction would exceed the cost of doing it in-house.
Section 9
Analyst's Take
Faster Than Normal — Editorial View
The best startups often reduce a large transaction cost. Find a market where exchange is painful — search is hard, trust is low, payment is clunky — and make it easy. That is the story of Stripe (payments), Airbnb (trust and search for lodging), and many marketplaces. The size of the opportunity is proportional to the transaction cost you remove.
Make vs buy is a transaction cost calculation. Do not compare only the vendor's price to your internal cost. Add the full cost of finding, contracting, monitoring, and the risk of hold-up or renegotiation. Often the "cheaper" external option is more expensive once transaction costs are included. Run the full comparison.
Your product can impose or reduce transaction costs for users. Every extra step, unclear term, or support runaround is a transaction cost you are adding. Every default, one-click flow, or guarantee is a transaction cost you are removing. The latter builds loyalty; the former builds resentment. Map the transaction costs your users face and reduce them where you can.
Platforms capture value by sitting at the point of high transaction cost. If you are the place where search, matching, or trust is solved, you can capture a share of the value that would otherwise be lost to friction. The defensibility of the platform is often the difficulty of replicating that transaction cost reduction.
Institutions and technology shift transaction costs over time. When a new technology or regulation lowers transaction costs in a sector, expect more outsourcing, more modular supply chains, or more market-based coordination. When transaction costs rise (trust collapse, legal complexity), expect more integration or long-term relational contracts. Position for the direction of change.
Measure before you assume. Do not guess whether your transaction costs are high or low. Map them: how much do we spend on search, negotiation, and enforcement? How much hold-up risk do we carry? Back-of-envelope estimates beat no estimates. Companies that run make-vs-buy or build-vs-partner decisions without quantifying transaction costs often choose wrong and only discover it later.
Section 10
Test Yourself
Is this mental model at work here?
Scenario 1
A company brings its customer support back in-house after two years with an outsourcer. The outsourcer's per-ticket price was lower, but renegotiations, quality disputes, and dependency on a single vendor made the total cost higher.
Scenario 2
A new app lets freelancers and clients sign standard contracts and release payment through escrow. Adoption grows quickly in a segment that previously relied on informal agreements and late payments.
Scenario 3
Two similar companies merge. Analysts say the merger is about 'synergy.' The main overlap is in procurement: the combined entity negotiates one set of supplier contracts instead of two.
Scenario 4
A retailer adds a 'buy now, pay later' option at checkout. Conversion increases. The retailer pays a fee to the BNPL provider.
Section 11
Summary & Top Resources
Transaction costs are the full cost of exchange beyond the nominal price: search, negotiation, enforcement, and risk. When they are high, firms integrate or form long-term relationships; when they are low, markets expand. Coase explained the boundary of the firm this way. Strategically: make vs buy depends on full transaction cost, and reducing transaction costs for customers is a powerful source of value and defensibility.
The origin of transaction cost economics. Coase asks why firms exist and answers: because using the market has a cost. Required reading for the theory.
Scott on how states and large organisations reduce transaction costs through standardisation — and what gets lost. Useful counterpoint to pure efficiency.
Coase's collected papers on the firm and transaction costs. Includes the 1937 and 1960 articles and later reflections. The canonical source.
Leads-to
Intermediation & Disintermediation
Intermediation adds a layer between parties; disintermediation removes it. The choice depends on transaction costs. When intermediaries reduce search, matching, or trust costs, they persist. When technology or design lowers those costs, disintermediation wins. Transaction cost logic explains when intermediation is worth it and when it is not.
Leads-to
Mechanism Design
Mechanism design asks how to design rules and institutions so that participants have incentives to reveal information and behave in a way that achieves desired outcomes. Transaction costs are part of the environment — mechanism design often aims to reduce or work around them (e.g. auctions that lower search and negotiation cost).
