The Replication Question
Context note: This post presents conceptual reflections. It does not describe, assess, or report on any specific organization. All examples are synthetic or composite and non-attributable. They are based on my ongoing fieldwork.
There is a question I keep being asked, usually after a talk, usually by someone who carries responsibility for a region: Can this be replicated?
The question is sincere, and I have learned to listen for what sits underneath it. The people who ask it often add a second sentence, more quietly: this is what we are supposed to be doing, and from where we sit, we cannot seem to do it. I have stopped hearing that second sentence as an admission of failure. It is a precise observation about instruments. This note is me working out, in the open, what an honest answer would be.
Why the question unsettles me
My first instinct was to treat the question as a compliment and deflect it. I no longer trust that instinct, because the question is not really addressed to the ecosystem I have been studying. It is addressed to my research.
When I laid out the research spine of this PhD, I made a promise: that the agreement layer of regional ecosystems, the enacted rules that decide who can decide, what counts as quality, and who absorbs the downside, can be made visible enough that regions detect fragility earlier and learn from what actually works. The replication question is that promise handed back to me by someone who needs it to be true. So I owe it a research answer, not reassurance.
What the fieldwork keeps showing
The fieldwork behind these notes includes a regional ecosystem I have been mapping from the inside, reconstructing, episode by episode, how cooperation between its organizations actually works. What I found there is a pattern the literature treats as rare and practice treats as nearly impossible. Relationships between organizations that deepen over the years rather than wear down into routine. Institutions that genuinely change, without any program instructing them to. A region quietly growing a capacity that none of its members owns.
I want to be careful here, because this is easy to romanticize, and I am wary of witness accounts that flatter the witness. So let me stay with the single mechanism that has taught me the most, and let me read it the way my own method requires: through the footprint it leaves, not through the story told about it.
It shows up like this. An organization in the ecosystem has long wanted to change something about its own work. The analysis exists. The motivation exists. The money could probably be found. What is missing is not a resource but a warrant: the standing to say, out loud, that the change is now legitimate. Then the relationship with a trusted partner supplies exactly that. The partner's standards, its visible seriousness, become the external ground on which the internal case can finally stand. The change happens. And in every formal record, the organization did it alone.
A constructed illustration, composited from many such episodes: a public institution wants to work differently with its youngest visitors—less program delivery, more open-ended making. Internally, the idea keeps losing: to schedules, to habit, to the absence of anyone with the mandate to say yes. A partnership reframes the same idea as meeting the partner's quality standard. Nothing new enters the building: no extra budget, no new method, no instruction from above. What enters is permission. The institution was not persuaded to change. It was authorized to do what it already wanted.
Reading the episode through my own lens
In the framing my PhD builds on, an agreement is not a document. It is the set of implicit and explicit terms in force the moment people interact, whether or not anyone knowingly consented to them (Ritchie-Dunham, 2024). What this episode shows is an enacted agreement doing economic work: permission moving from one organization to another and leaving observable residues behind. An initiative that suddenly proceeds. A sentence that becomes sayable. A mandate nobody had to issue in writing.
And here the discussion with myself begins in earnest, because two things in this episode press against my own vocabulary.
The first: authorization cannot be self-issued. An organization can allocate its own budget, train its own people, and restructure its own processes. It cannot lend itself external standing. Whatever else moves through an ecosystem relationship, this particular good exists only in relation. It is produced between organizations or not at all.
The second, I can make visible with a simple test. Suppose the relationship ends, and suppose both sides keep everything they own: budgets, staff, competencies, contracts, accumulated knowledge. On paper, nothing has been lost. What has been lost is a sentence. "We are doing this because the partnership asks it of us" can no longer be said. Nothing was removed from either organization because the thing that disappeared was never in either.
Part of a region's capacity to change lies in the relationships between its organizations, not within any of them. Not metaphorically, the way goodwill is said to live in a brand, but operationally: the capacity exists exactly as long as the relation exists, and it has no organizational address.
I notice what this does to my own conceptual apparatus. My vocabulary was built for the negative case: agreement debt, the hidden cost that accumulates when coordination is fragile. What I met here is the mirror image, a surplus accumulating in the same layer, and I do not yet have a name for it that I trust.
The research neighborhood does not quite have one either. Strategy scholars have shown that relationships hold value that no single organization could produce alone (Dyer & Singh, 1998; Nahapiet & Ghoshal, 1998). What I kept seeing is one step stranger than value. These relationships held permission, a lateral form of what institutional theory calls legitimacy (Meyer & Rowan, 1977). Value can be appropriated, and access can be substituted. Permission that a relation issues ends when the relation ends.
So: can it be replicated?
The honest answer my own framework allows is narrower than the question wants, and more hopeful than it fears.
Not by copying an organization, because the thing worth copying is not an organization. What can travel are conditions: relationships treated as infrastructure rather than as a means to projects; agreements taken as seriously as budgets; and some way for the system to see itself. What cannot travel by decree is the layer itself. It is grown in place, or it does not exist.
Everything in me that wants to say more than that runs into a rule I set early in this PhD and refuse to break: no recommendations. Not because restraint is comfortable, but because advice would displace the one thing that seemed to matter most in the field. The growing, as far as I can see, is unglamorous. Sit with people long enough to reconstruct concrete episodes of cooperation. Trace what actually moved between organizations. Draw the causal structure that emerges. Then hand the drawing back to the people in it, as a hypothesis, to be corrected. The moment that keeps surprising me is the correction. When people repair the map of their own system, the system has begun, for the first time, to see itself. I suspect that this, more than any finding, is the intervention. And it is nothing a decree could produce.
Why can't the instruments do it
Which brings me back to the quiet second sentence: we are supposed to be doing this, and we cannot. From where public instruments sit, I think that is structurally true rather than personally true. Instruments route resources toward addresses: toward what can be owned, audited, and attributed. Programs strengthen the parts. The layer between the parts, where a share of the region's capacity to change is actually held, remains outside the transaction, because nothing there can receive funds, sign for them, or account for them. A region can be well funded at every address and undersupplied precisely in the layer that authorizes change.
This is not the failure of any particular board or administration. It is a property of instruments that requires owners. And the people inside those instruments are often the first to sense it. That, I have come to think, is what the replication question is actually saying.
What this note taught me about my own inquiry
I began this note thinking I was answering a question from outside my research. Ending it, I think the question was inside all along. My PhD thesis states that the layer between organizations can be registered through agreement footprints: observable residues, patterns rather than opinions, evidence that does not need an owner to exist. If that bet holds, then what a region cannot copy, it could at least see. And what it can see, it can begin to deliberate about.
What remains unfinished is the translation. An earlier note argued that standard measurement systems are calibrated to the wrong unit: they assume value can be attributed to a unit that holds it, and relations do not submit reports. Between a footprint in the field and a line, an investment committee can responsibly act on conservative translation without fake precision. That is the part of the promise I have not yet kept. It is where this inquiry goes next.
References
Dyer, J. H., & Singh, H. (1998). The relational view: Cooperative strategy and sources of interorganizational competitive advantage. Academy of Management Review, 23(4), 660-679. https://doi.org/10.5465/amr.1998.1255632
Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: Formal structure as myth and ceremony. American Journal of Sociology, 83(2), 340-363. https://doi.org/10.1086/226550
Nahapiet, J., & Ghoshal, S. (1998). Social capital, intellectual capital, and the organizational advantage. Academy of Management Review, 23(2), 242-266. https://doi.org/10.5465/amr.1998.533225
Ritchie-Dunham, J. L. (2024). Agreements: Your choice. Vibrancy Ins, LLC.
