Developed by Paul A. David (Clio and the Economics of QWERTY, 1985), W. Brian Arthur (Competing Technologies, Increasing Returns, and Lock-In by Historical Events, 1989), and Douglass C. North (Institutions, Institutional Change and Economic Performance, 1990).
Traces how an early, partly accidental choice became locked in through network effects, learning, and switching costs, and estimates what it would take to break free.
What holds a hospital to its old records system, and what it would take to break free
Each lock is what the hospital would pay to leave, in millions over five years; each gain is what a better system would return. As set, the locks outweigh the gains by $2 million, so the hospital stays. Then move the new interoperability rule, which removes part of the interface work, and see whether the case to switch tips. Illustrative.
An example
A 300-bed community hospital had run the same electronic records system for fourteen years. Clinicians disliked it, and every proposal to replace it died at the budget committee. The new chief operating officer asked why before anyone wrote another.
The trail went back to a 2011 choice: the hospital had bought the system its largest physician group already used, mainly to save on one interface. Since then three locks had formed. Some 2,400 staff had learned the system and its workarounds. Thirty labs, pharmacies, and clinics had built custom interfaces to it. And the 2019 renewal carried a steep exit fee. Priced over five years, the locks came to about $11 million against $9 million in gains, so each committee had been right to say no.
The analysis also found the break. A state rule taking effect in two years would require standard interfaces for every records system, removing most of the interface lock. The hospital set its switch for the year after the rule and negotiated the exit fee down at the next renewal.
Why it works
Being stuck with an arrangement everyone agrees is worse feels irrational, and the usual response is another argument for the better option. Path dependence explains why the argument keeps losing: an early, often accidental choice gets reinforced by everything built on top of it, until leaving costs more than the improvement is worth. The arrangement survives because of its history, not its merits.
Naming and pricing the locks turns frustration into something workable. Each lock has a different remedy: learning can be rebuilt with training, network ties need a standard or a bridge, contracts expire or can be bought out. And locks have events that weaken them: a new rule, a technology shift, a contract renewal, a merger. The analysis says when switching becomes worth it, which is often a date, not a debate.
The thinking behind it
Paul A. David Paul A. David, an economic historian of technical change, made path dependence a working idea in economics with Clio and the Economics of QWERTY (1985). His account of the typewriter keyboard argued that small historical events, technical interrelatedness, and increasing returns can lock an industry into a standard that persists after better alternatives appear. The case made path dependence a standard explanation for why markets do not always settle on the best design.
W. Brian Arthur W. Brian Arthur's Competing Technologies, Increasing Returns, and Lock-In by Historical Events (1989) showed how small chance events early in the adoption of competing technologies can lock a market into one of them, even an inferior one, once each adoption makes the next more likely, so that which technology wins depends on the order of early adoptions. He collected the work in Increasing Returns and Path Dependence in the Economy (1994).
Douglass C. North Douglass C. North, in Institutions, Institutional Change and Economic Performance (1990), defined institutions as the rules of the game in a society and showed how they persist and constrain later choices. His version of path dependence explains why inefficient arrangements, from property rules to organizational habits, can stay in place long after everyone sees their cost. He shared the 1993 Nobel Memorial Prize in Economic Sciences.
How to run it
- Name the stuck state and the better option. Write what everyone agrees is worse and the alternative they would choose if starting fresh.
- Trace back to the fixing choice. Find the early decision that set the course and note how accidental it was, which helps separate history from merit.
- List the locks. Sort them into network effects, learning, expectations, and switching costs such as contracts and sunk assets.
- Price each lock. Estimate what leaving would cost through each one, over the same horizon as the gains.
- Find the events that weaken a lock. Look for rule changes, standards, renewals, new technology, or mergers that would cut a lock's price.
- Decide to stay, hasten the break, or route around it. Live with it, time the switch to the weakening event, or build a bridge that lets the better option coexist.
Where it goes wrong
The common failure is arguing merits against a lock. A better proposal goes to committee for the fifth time and loses for the fifth time, because the locks were never priced. It shows up as frustration with decision makers who are behaving rationally. Price the locks first.
A second failure is missing the lock being created now. A decision today, a vendor, a data format, a contract term, can fix the path for a decade. When a choice touches many other systems or people, ask what it would cost to undo in five years. A third is treating every unpopular arrangement as path dependence. Sometimes the old system persists because it is still the better fit; check the gains as hard as the locks.
When to use something else
- Sunk Cost Reasoning. When the pull to stay comes from money already spent that cannot be recovered, and not from costs still to come.
- Network Effects. When the main lock is that the arrangement's value grows with the number of users, and the work is growth or competition strategy.
- Hysteresis. When a system does not return to its old state after the cause is removed, and the question is why it stays shifted.
Where it sits in SOLVE eX
Go to the source
- Clio and the Economics of QWERTY, American Economic Review 75(2), Papers and Proceedings, 1985
- Competing Technologies, Increasing Returns, and Lock-In by Historical Events, Economic Journal 99(394), 1989
- Institutions, Institutional Change and Economic Performance, Cambridge University Press, 1990