Path Dependence - OrgRiver

Early choices can reshape the options available later, because learning, investments, standards, and relationships accumulate around the path already taken, making alternatives progressively less attractive relative to the installed path, even if the original advantage was small or accidental.

3 min read

What Is It?

Economists studying technological adoption in the 1980s made the QWERTY keyboard layout a famous illustration of path dependence. Paul David argued that an early technical standard could become entrenched as users learned it, manufacturers standardized around it, and complementary investments accumulated around the installed base. The details of the QWERTY case specifically, and whether an alternative layout would actually have been superior, have since been debated, but the broader mechanism doesn’t depend on that particular example holding up. The mechanism doesn’t require the early choice to have been the best one, or even a particularly deliberate one. Once a path is chosen, learning, complementary investments, standards, and relationships can begin accumulating around it, and those reinforcing effects make alternatives progressively less attractive relative to the installed path, regardless of whether a different path would now be better.

Why Does It Matter?

Organizations accumulate technical, structural, and cultural choices the same way, an early tool selection, an early org chart, an early process, that made sense at the time and then became progressively harder to change as more was built on top of it. A database chosen early because it was fast to set up can still be running the business a decade later, not because it’s still the best option, but because everything built since assumes it’s there. A reporting structure created to solve one team’s problem years ago can persist long after the problem changed, because untangling every process and relationship built around it costs more than anyone wants to spend fixing something that technically still works.

This creates a specific organizational trap: at any given moment, staying on the current path can be individually rational, since the switching cost is real and immediate while the benefit of switching is often diffuse and long-term, even while the organization would clearly be better off having never gone down this path in the first place. Nobody has to make an irrational decision for the organization to end up on a path it wouldn’t choose if it could start over.

What Changes Once You See It?

You start distinguishing “this is still the best choice” from “this is just the choice we’ve built the most on top of,” since path dependence can make the second one look and feel like the first without actually being it.

You start treating early decisions, especially on things likely to be built upon extensively, with more deliberate scrutiny, since the real cost of an early choice isn’t the choice itself, it’s every future decision that will treat it as a fixed constraint.

You also start asking whether lock-in is strengthening over time, since when complementary investments keep accumulating around the current path, postponing a switch can make an already difficult transition still harder, though that isn’t automatic, sometimes the existing system depreciates or a new interoperability layer lowers the cost of switching later instead.

Common Misunderstandings

It isn’t a claim that every long-standing choice is a mistake. Sometimes the early choice really was, and still is, the best option, path dependence describes why change becomes harder over time, not that change is always warranted.
It isn’t the same as Structural Inertia, which is about organizational resistance to change in general. Path dependence is more specific: it’s about how a particular early decision generates accumulating switching costs that constrain later choices, even absent any deliberate resistance to changing.
It isn’t the same as the Sunk Cost Fallacy. Sunk costs are past expenditures that shouldn’t affect a forward-looking decision, path dependence concerns real present-day constraints created by past decisions, compatibility, retraining, coordination, dependencies, that genuinely affect today’s alternatives.
It doesn’t mean small early decisions should be treated as irreversible from the start. Many early decisions genuinely are easy to reverse while young, the risk grows specifically as more gets built on top of them, which is why timing matters more than the decision’s original size.
It isn’t limited to technology choices. It applies just as much to organizational structure, cultural norms, vendor relationships, and process design, anywhere later decisions build on and reinforce an earlier one.

Diagnostic Question

If we were starting from scratch today, would we choose this path again, and if not, what has accumulated around the current path that keeps us on it?

Explore Further

Field Notes

None yet.

Related Field Guide

Origin

Paul David, “Clio and the Economics of QWERTY” (1985), American Economic Review, using the QWERTY keyboard layout to illustrate how early technology choices can persist through accumulating switching costs; the broader concept was further developed by economist W. Brian Arthur’s work on increasing returns and lock-in.