BUSI 103 - Introduction to Business (Chapter 6)
August 3, 2026
What should your business do itself, and what should it let somebody else do better - and how do you know when to redraw that line?
Read time ~26 min - ~3,830 words - problems ~40 min
Why this matters. Every business is full of make-or-buy calls - what to hire in, what to hand to specialists, what to keep because it is the thing that makes you special. Draw the line well and you gain focus and cost advantage; draw it badly and you buy complexity, distraction, and someone else’s problems.
What you’ll be able to do.
The big ideas.
Key terms. boundaries of the firm, transaction costs, bundle of activities, make-or-buy decision, vertical integration, outsourcing, learning curve effects.
A mid-sized specialty coffee roaster built its reputation on the beans: it sources and roasts them itself, because that is what customers pay for. Shipping is a different story. The roaster handles its own nationwide shipping today, and it is debating handing the whole thing to a third-party logistics (3PL) provider. Here is the wrinkle: the 3PL will charge more per shipped order than the roaster’s current in-house shipping cost. But the 3PL is known for higher reliability, better tracking, and fewer late deliveries - and the roaster expects fewer customer complaints and less management time spent firefighting.
Paying more per order, to have somebody else do a job you already know how to do. Is that a deal a sensible manager takes? Hold the question - this chapter is about how to think it through.
Firms exist because using the market is not free - transaction costs decide what you hire in and what you buy.
When we talk about boundaries of the firm, it’s important for every business. We kind of assume we know what we want to do, and that’s actually a decision, right? The boundaries tell us what we’re going to do, what other people are going to do better. If you think about every single business making things from scratch - from atoms, rather - they’re deciding what to buy, and it might seem natural that we’re buying raw materials. You can ask yourself: why don’t we make those raw materials? We might have our onboard accountant, or we might hire a temporary accountant or a fractional person to help us do that. The decision on what we do ourselves versus what we rely on other people to do, or perhaps technology providers provide for us, are all compositional decisions.
This was brought out in some work done by an economist named Ronald Coase on the whole idea of a theory of the firm. It’s a very simple question, but it’s actually quite profound. And the question is: why do firms exist at all? If markets are efficient, every single time you need something, you just go to the market and it would be there. Just contract for every service that you need. Instead of hiring an employee, creating departments, doing this yourself, why not just go to the market?
The answer, in his mind, was transaction costs. It costs us something to do that: to go into the market searching for something, or to be on the market providing, let’s just say, a particular service. Let’s just make an extreme example: going out there helping people stock a supply closet. The thing is, people would have to learn that they have to do it themselves. They would have to find you. They would have to find out the details of what it is like to work with you about supplying their office supply closet. Is that going to be worth it? Should they do that themselves? Should they ask you to do it? Is it going to be cheaper? Is it going to be more expensive?
This is a lot of questions. The market’s not free. It requires time for discovery, you have to coordinate, you have to negotiate the terms, and because they’re not really on your team, you’re working with them through a transaction. There’s a risk: what if they don’t do what you want? What if they’re not quite good enough? How do you interact with them? Working with the market can be a little bit more tricky than working with an employee that is your own.
So this is why we have firms. To reduce this transaction cost, we hire somebody and we say, “Hey, basically, you’re on our team.” Different things may come up, but your job is to be on our team. Think about how to help us, and we might change the terms of what we need you to do. In general, you work for us, and we can kind of be a little bit looser with very specific things. We don’t have contracts for a single task and behavior - we internalize those costs of using a market when you bring them into a firm.
We outsource when the market’s more efficient - if there’s something that’s very clear what we need done and the expertise is out there, rather than hiring it, developing it, maintaining it ourselves. We see that happen in a lot of different cases. We don’t do our own dry cleaning. We don’t get a whole selection of shoes when we wanna get shoes. We go out to a store, and there’s a retail store whose whole job it is to shop for different shoes, provide an assortment so we can come in and do that. There are so many parts of our lives where we use the market, and we just assume that it’s there, but turns out it’s a complete decision on how markets are structured, whether or not we use the market or do things for ourselves.
Bottom line: internalize when the market is expensive to use; go to the market when it is the more efficient way to get the thing done.
A business is a bundle of linked activities - and not one of them is in-house by default.
This same kind of question exists for a business when it comes to the boundaries of the firm. A neat way of thinking about firms is as a bundle of activities. There’s sourcing materials, manufacturing or delivering a service. You have to brand and market, tell people what it is that you do, sell, distribute, work with customers, and afterwards provide customer service. There’s a whole set of support activities around billing and IT and making sure everything kind of moves, and hiring people and training people and keeping things going. This whole thing is like a bundle. It’s a system of a business, and not all of these activities have to be done by you. You can hire other people to help you do this, or, in some cases, carve something out and do it completely.
Bottom line: drawing the boundary means deciding, activity by activity, which pieces of the bundle are yours and which belong to the market.
Firms genuinely differ in what they are good at - and those asymmetries decide who should do what.
Why are some firms better at certain activities instead of doing them yourself? They might have knowledge or expertise - domain-specific know-how. For example, doing taxes. A lot of companies hire other people to do taxes because someone else is better at doing that. Because they do a lot of different taxes for different people, they get good at doing that. It becomes efficient to have them do it. It’s basically just Ricardian comparative advantage. They might have scale economies: this is something that we want to have done, but when you do it at scale, you can do it cheaper or you can do it better. Learning curve effects are kind of like scale - the more we do it, the better we get at it.
It could be coordination or system design. If there is a task you need that requires coordinating with a lot of different people, and they’re good at doing that - for example, a supplier can call on all these different factories to produce shoes and coordinate that they do as an ongoing business. You don’t have to do that all yourself. If you were to replicate what they do, it would be very costly for just a handful of pairs of shoes, whereas for them, they’re selling so many shoes to so many people that this is worth it.
And there are relationships and culture. It takes some time to be in connection with a lot of suppliers. Let’s just say there’s somebody who runs a store, and they talk to a lot of the suppliers of different foods, of groceries, of meats, of vegetables. They talk to farmers, talk to people who make cheese, instead of you talking to every single one of them. They just talk to all of them, and they bring it all into one market where you can have an assortment.
These differences create asymmetries in who’s best at doing that and who should do it. If a supplier is better, you should outsource to them - provided that, because there’s a cost to working with suppliers, the cost of that transaction is less than doing it yourself. This is the essential question of boundaries of the firm.
Bottom line: the boundary follows the asymmetries - who is better at the activity, by enough to cover the cost of working across the boundary.
The boundary is a standing managerial decision - make or buy, asked and re-asked as conditions change.
Boundaries of the firm are a managerial decision. It’s not a one-time decision. We’re constantly looking at it. We always ask: What do we do uniquely well? Where are they better than us? Where can other people do this more efficiently, more flexibly? How expensive is it to coordinate with them? As different things change, we might change our minds about where we draw those boundaries of the firm.
This is the classic make-or-buy decision - basically the frontline piece of boundaries of the firm: do we produce this input or perform this function ourselves, or do we contract it out? The answer kind of depends on four key questions:
That last one is a good question, and it deserves its own section - hold it for 8.6.
Bottom line: make-or-buy is a question about advantage, not just price - and it never stays answered for good.
Keep what is central to your advantage in-house; hand off what somebody else does better.
Here’s what you want to keep inside. Vertical integration is getting involved in greater aspects of the value chain - taking more ownership over that. If you need tight control over the quality and the timing of the process, you might want to own that. If the market’s just unreliable - there are just not many good suppliers out there that you can work with - you might want to do things yourself. You should do it if you are big enough to achieve economies of scale: rather than having somebody outsource, you’re big enough to justify having and maintaining that capability in-house. You should do that if you are actually smart at this and possess unique knowledge or IP for doing this well, and if this task is just central to your brand.
Here’s an example - and it should sound familiar. There’s a specialty coffee roaster who controls the sourcing and roasting of the coffee beans because that’s central to their experience. If they really make their reputation around the quality of the coffee and they think it’s just tied to a lot of these things - where you source coffee and how you roast it - these are things they don’t want to ask other people to do. This is what makes them special, and that’s why you might want to insource that.
Whereas if coffee is just one other beverage and you’re actually a full-scale restaurant, it’s just a part or a very small part of what you’re offering. You wouldn’t mind outsourcing this because it’s not that special what you do. Somebody else is probably thinking about this more, doing this better than you can, and can supply it to you cheaper than you can doing it yourself.
Notice where that leaves our roaster from the top of the chapter: the beans stay inside because they make the brand. Shipping is the piece of the bundle where somebody else might have the edge - which is exactly why the 3PL call is the hard one.
Bottom line: keep the activities that make you special; let specialists carry the rest - and know which is which.
Outsourcing does not make costs disappear, and a supplier’s margin can never truly be bought.
Some people think somebody else is making a lot of profit. In their supplier review, it’s easy to say that those people are making a lot of money off us, and we shouldn’t let them do it. We could make that money instead. It comes to this question when it comes to vertical integration: hey, our suppliers make a lot of money. Can we make that money instead - in so many words, buy their margin?
The thing is, integration isn’t free. You take on this job, and you might find out they’re earning this much, and there’s a very good reason. It’s complex to manage the stuff. It’s going to incur new cost. Coordinating this stuff is burdensome, and now we’re distracted from what our core priorities are. We are now focusing on this thing, which is supposed to be ancillary - and it might be that being really good at this thing is kind of culturally different from what we do. If you’re running a medical supply store, you want to be really efficient, but there’s this thing that takes a lot of careful attention and quality control for making this product. It’s hard to be really interested in details and moving fast. It’s helpful that there’s a whole other organization that really focuses on doing that well, and it’s difficult to do both kinds of things and have two different mindsets in the same place.
If we decide to do it ourselves, we don’t want them to earn the margin - we want to earn this margin, and we want to vertically integrate. What we might find is that this is not worth it. That margin comes at a cost, and by the time we capture it, we’ve kind of wasted so much trying to support this thing. The same way, if you outsource, you might gain flexibility - this is something I don’t have to do anymore - but then we don’t have control over that. We’re not sure how it’s done. We’re worried about quality control, and it turns out all the extra effort we have to incur dealing with a supplier is worse than doing it ourselves. We always have to focus on the total cost of this arrangement - which, by the way, is kind of high-stakes and kind of important. You have to think about it more completely.
Now, you can’t really just get rid of cost by giving it to somebody else. You can’t outsource costs. The reason for this is that if it’s costly, no business can survive helping with this without incurring the cost themselves. This is eventually going to come back to you in the form of some price and some markup.
At the same time, you also can’t capture someone’s margins. If they’re making a margin and you think, “Hey, my current suppliers make a lot of money off me. I don’t want them to do that,” and you start getting into that business, just keep in mind you’re also going to incur all the extra costs of doing that. It might be, if you’re an efficient producer, that it’s going to be more expensive for you to backward integrate and do what they do, because there are some things that they do that you’re not aware of. It could also be that you will incur some costs from the split attention. Now that you’re worried about multiple fronts and you have to focus on this thing as well, you just end up not being as good as they are. The cost can come back in terms of you forgoing opportunities, just not executing as well, because now you’re trying to do two things. The margins from others can never truly be bought.
There’s one more layer: the hidden costs and hidden benefits. It’s one thing to say we’re just saving costs - they can do it cheaper - but what will that mean for our freedom? Our freedom to make choices, our freedom to react to markets and opportunities, our freedom to innovate. These are things that are harder to account for with cost, but it’s something that we need to think about.
Bottom line: price the whole arrangement - the coordination, the distraction, the freedom you give up - not just the line item you think you’re saving.
Boundaries move when technology changes the economics of doing it yourself versus using the market.
Technology can shift the boundaries. There are things that we used to hire other people to do, and with the emergence of some technologies, that could change where the boundaries of the firms are drawn.
For example, it used to be that people used a lot of computers and had big main centers, hired a lot of people, and had a lot of rooms full of computer equipment. Now that we can do a lot of things in the cloud, someone else can centralize that. A lot of small firms that need computing resources that they can’t maintain themselves might come into existence. A lot of tasks that used to be handled only by people who could afford to manage a lot of computer stuff - now different firms can emerge, because they can kind of separate those two things. Somebody can deal with maintaining all this computer equipment and hardware, and other people are focusing on applications that run on them.
As technology changes, the environment changes. It might shift the whole logic of whether we want to make or buy, and whole new industries, whole new services, can emerge on the backs of that technology development.
Bottom line: when integration costs fall, the case for outsourcing gets stronger - so keep asking what has changed in the environment that shifts the logic of make versus buy.
Real firms redraw boundaries in every direction - a sliver of ownership, the whole stack, or handing off the core.
A couple of examples. One of them is Starbucks. Starbucks was selective about upstream integration. In 2024, they acquired some farms in Guatemala and Costa Rica. Now, these farms only supply a very small amount - like 3% of its total beans. Why do we care about this? It’s not about fully integrating and controlling the whole process of getting coffee beans. It was trying to learn something about quality control, about sustainability practices, and about learning more about suppliers by getting into that business. Firms sometimes are going to draw some vertical boundaries not because you’re fully going to integrate and get completely into that business - just having some ownership in that area allows them to get better at the job that they do, such as shaping standards, branding, and learning what its suppliers have to do.
Tesla’s an example of a lot of full-stack integration. They have battery production, and they also design vehicles. They designed software, and they even do their own retail and their own servicing. This is a new industry, and succeeding here takes a lot of learning. A lot of automakers are outsourcing major systems - the brake system, the tires, the interior, and the entertainment system - and they’re primarily just really focused on the drivetrain. But Tesla owns everything kind of end-to-end. It lets them innovate faster, and they control more of the customer experience. When there’s some complexity around how, let’s say, the hardware and the software interact, owning the whole end lets them be a better innovator here. Vertical integration can be helpful when we’re still learning about a whole system - we don’t know quite how it works - and when speed and coordination matter a lot.
The general rule, when we think about it, is: don’t outsource things that are critical to your core competency. But look at how Apple and contract manufacturers like Foxconn have partnered. Apple has a lot of design, and they do a lot of marketing and services, but a lot of their manufacturing is outsourced to places like Foxconn. Foxconn has the scale, labor, and specialized processes. It got to the point where they’re actually just as good, sometimes better than Apple, at some of these manufacturing processes. The lesson here is that even things that are considered core can be learned to be outsourced - but this comes with some costs: the cost of control, and the fact that Apple is now very dependent on Foxconn. Some people debate whether that’s just the right call. This is just to show that there are cases where even some very fundamental things, like core manufacturing, can in fact be outsourced, and some companies like Apple have done it.
Bottom line: Starbucks bought a sliver, Tesla owns the stack, and Apple handed off the factory - three different answers to the same Coase question, each set by its own context.
So does the roaster hand its shipping to the 3PL? Run the question the chapter has been asking. In the end, this is what the business side has to decide: what do we do best, and which do we let other people handle? What are our capabilities, and where do we want our capabilities to be? Shipping is not what makes this roaster special - the beans are. The 3PL’s sticker price is higher per order, but the roaster is not just buying shipping; it is shedding the coordination, the monitoring, the firefighting - the transaction costs of running a function the 3PL runs at far greater scale and reliability. That is exactly the trade Coase described, running in reverse: use the market when the market, all-in, is cheaper than managing it yourself. The one honest warning from 8.6 still applies: the roaster is not getting rid of its shipping costs - it is trading its cost structure for the 3PL’s, plus the 3PL’s margin - so the deal only works because the 3PL is genuinely better at the job. And the decision is not forever. Keep in mind, the boundaries of firms are the result of a lot of different contextual features - what the state of technology is, what your capabilities are, what other people’s capabilities are. All of these things are open to changing and evolution, and the boundaries of firms are always being redrawn and negotiated.
6-1 The Buckle Temptation. [LO2, LO3] An outdoor-gear company buys aluminum buckles from a single supplier and notices, in its annual supplier review, that the supplier earns a healthy margin on every order. The CEO proposes bringing buckle production in-house “to capture that margin ourselves.” (a) Walk the proposal through the four make-or-buy questions from 8.4. (b) Name three costs of integrating that would never appear on the supplier’s invoice. (c) State the conditions under which integration would make sense here.
6-2 The Boundary Audit. [LO1, LO2] Pick a real business you know well - a campus cafe, a local gym, a family business. (a) Map its bundle of activities: list at least six, from sourcing through customer service and support functions. (b) Mark each one in-house or outsourced, as best you can tell. (c) Pick two of those placements and justify them using transaction-cost logic. (d) Identify one activity you would move across the boundary, and argue why.
6-3 The Data Center Question. [LO4] A regional insurance company has run its own data center since the 1990s - servers, staff, a dedicated building. Leadership is debating a move to a cloud provider. (a) Explain what has changed in the make-or-buy logic since the data center was built, using Coase’s framework. (b) Name two new risks or dependencies the cloud move creates. (c) The CIO says, “we’d be paying someone else’s margin on computing we already own.” Respond.
© 2026 Eric Lin. All rights reserved. This chapter is provided for students in BUSI 103 - please do not repost or redistribute without permission.