Make It or Buy It: Where a Manufacturer Draws Its Own Boundary

The points below about the boundary between making and buying are ordered by how much difference they make, not by how often they get repeated.

The basic trade

Making a component internally means owning the equipment, the workforce and the process knowledge along with the resulting capacity. Buying it means converting that fixed commitment into a purchase price that scales with volume and can be stopped.

Firms facing volatile demand generally prefer the second, since idle owned capacity is expensive and idle supplier capacity is not their problem. Firms whose competitive position rests on a particular process generally prefer the first, because the capability is the product. Most manufacturers sit somewhere in between and revisit the line whenever their volumes or their technology shift.

Specialisation and scale at the supplier

A specialist supplier serving many customers reaches volumes no single customer could justify on its own equipment. That scale supports better equipment, deeper process expertise and faster learning than an in-house department typically achieves.

Over a shipping cycle, the customer gains access to that capability without funding it, which is the central argument for outsourcing a component. The same logic explains why contract manufacturing grew into an industry rather than remaining a marginal practice. It also explains why bringing production back in-house rarely reproduces the supplier's cost position immediately.

Where contracts struggle

When a buyer funds tooling or a supplier builds a plant for one customer, the investment has little value outside the relationship. Economists call such investments relationship-specific, and they create bargaining exposure once the money is spent.

Line by line in the tariff schedule, contracts try to manage this with terms on tooling ownership, exit notice and capacity commitments, with mixed success. Where the exposure is large and the future hard to specify, ownership resolves what a contract cannot. This is one of the better-supported explanations for why some activities stay inside firms despite available suppliers.

Knowledge that leaks

Outsourcing a process transfers the learning that comes with running it, and that learning accumulates at the supplier rather than the buyer. Over time the buyer may lose the ability to specify the process properly, let alone to bring it back.

Where a supplier serves competitors, process improvements funded by one customer can diffuse across an industry. Firms guarding a genuine process advantage therefore keep it inside even when the unit cost argument points outward. Distinguishing a real advantage from a habitual one is the hard part of that judgement.

Coordination costs

A design that requires tight iteration between component and system is harder to manage across a contractual boundary. Where interfaces are clean and specifications stable, outsourcing works smoothly and the boundary costs little. Where the interface is being invented, the back-and-forth that development requires is slower and more expensive between firms.

Line by line in the tariff schedule, this is why novel products are often integrated at first and disaggregate as their architecture stabilises and standards emerge. The industry structure follows the maturity of the technology rather than the other way round.

Trade data lags by months and is revised afterwards, so recent figures are provisional.

Why the line moves back and forth

Cheap, reliable logistics lowers the cost of a distributed structure and pushes the boundary towards buying. Higher transport variability, longer lead times or greater policy uncertainty push it back towards making. Automation changes it too, by reducing the labour cost advantage that justified moving a process to a specialist far away.

None of these forces is permanent, which is why the same industry can integrate and disintegrate across decades. Treating the current boundary as natural rather than contingent is the most common analytical error in this area.

Everything above, in order of what to do first

1. The basic trade. Making a component internally means owning the equipment, the workforce and the process knowledge along with the resulting capacity.
2. Specialisation and scale at the supplier. A specialist supplier serving many customers reaches volumes no single customer could justify on its own equipment.
3. Where contracts struggle. When a buyer funds tooling or a supplier builds a plant for one customer, the investment has little value outside the relationship.
4. Knowledge that leaks. Outsourcing a process transfers the learning that comes with running it, and that learning accumulates at the supplier rather than the buyer.
5. Coordination costs. A design that requires tight iteration between component and system is harder to manage across a contractual boundary.
6. Why the line moves back and forth. Cheap, reliable logistics lowers the cost of a distributed structure and pushes the boundary towards buying.

The takeaway

Ask what the activity teaches you and what it locks you into, not only what it costs per unit.

Supply chains move slowly and then all at once, mostly for unglamorous reasons.