Connection as Design Principle — Statecraft

A note for international readers

This paper describes an intervention in the Dutch jeugdzorg (youth-care) system, which since 2015 has been the responsibility of municipalities rather than central government or insurers. Each of the 342 Dutch municipalities holds the budget for prevention, ambulatory care, residential placement and youth protection within its boundaries, with referrals coming through general practitioners, certified institutions and the municipality’s own access teams. The decentralisation was meant to bring care closer to families and reduce volume; in practice it produced structural deficits in most municipalities, including the one described here. Two terms recur and may be unfamiliar: SKJ refers to the professional registration body for social workers in the youth-care field, equivalent in function to the Health and Care Professions Council in the United Kingdom; an interim manager in the Dutch public sector typically holds executive authority for a defined period to lead a turnaround, structurally different from a management consultant. Where helpful, footnotes provide further context.

The test that changed the work

The brief was a twelve-million-euro saving on a youth-care budget that, without intervention, would have grown further. The approach set in the opening conference did not centre on cuts. The test agreed there was this: it is acceptable for a family to require considerable care, provided that family and professional are both satisfied with the approach. Three years later the saving had been delivered, with a council that stayed on side and a portfolio holder who continued to carry it politically. The financial outcome followed from the substantive work, rather than the other way around. That distinction is rarely as clean in public administration as it was here, and it explains why the approach held where comparable interventions elsewhere tend to drift.

What became visible in the social-domain monitor over subsequent years may be more interesting than the figure itself. The stacking analysis showed a fractal distribution: roughly ten per cent of households used half the budget, and within that ten per cent the same pattern repeated, with a fraction of one per cent accounting for a quarter of total spending. This is not a Pareto curve. It is a power law, the kind of distribution that characterises complex social systems in which no single scale is the “real” scale. An approach designed around averages would, by definition, have missed its target. The leverage lay not in proportional effort across all families, but in an intervention logic that could operate on multiple scales simultaneously.

The institutional backdrop

Anyone observing the Dutch youth-care system since 2015 from a distance can identify the pattern described in an earlier Statecraft paper as institutional dissociation. The provider extending a placement is not acting in bad faith. They are completing an order form the municipality has issued. The municipality issuing the order is processing an indication set elsewhere. The professional setting the indication follows clinical guidelines from their professional body. The guidelines derive from internationally codified diagnostic categories. No single link sees the total output. Everyone sees their own link working correctly. The growth in volume is not a decision. It is an artefact of a chain whose sum is held by no one.

In this light, the conventional reform logic is itself part of the problem. Stricter oversight frameworks lengthen the chain. Harsher sanctions on documented errors raise the incentive not to document. Personal liability for senior officials produces defensive organisational cultures in which no one decides and all responsibility migrates to collective bodies. What has been demonstrated in healthcare, education and housing associations applies equally to the youth-care system: more architecture does not solve dissociative architecture. What does work, or more accurately, what works less badly, is connection. Not as therapeutic concept, not as participation pledge, but as a constructed setting in which substance carries more weight than procedure.

What happened in the municipality at the centre of this paper is, in that light, not a successful budget cut but a deliberate dismantling of several dissociations at once. Eleven interventions, none a magic bullet on its own, together producing a recovery that registered in the figures. The order in which I describe them below reflects not a methodology but a pattern that revealed itself gradually.

The eleven connections

The conference as place. The work began with everyone in the room: providers, referrers, neighbourhood-team workers, councillors, the portfolio holder, family representatives, policy officers. One space, with the financial and substantive reality jointly on the table. This is not a meeting culture. It is dissociation undone at a level that institutional architecture normally cannot reach. The setting in which the entire chain looks each other in the eye and must weigh the sum is, in conventional administrative practice, absent. By constructing it deliberately, the first link of substance came into being.

The over-commit with reinvestment. The conventional reflex when faced with a budget shortfall is to cut the apparatus. The visibility of staff costs makes them the first target, and the political reward for “less bureaucracy” is sizeable. In this brief I reversed the reflex. Against the requested twelve-million saving I offered fifteen, on condition that I could deploy three million in additional resources to recruit staff, build the data system and resource the core team. The argument I used was arithmetically irrefutable. Even if every officer in the social domain were dismissed, the saving would not cover the deficit. The care budget in a municipality of this scale is many times larger than the staff capacity that steers it. Anyone cutting the apparatus to escape a care deficit is operating in the rounding error of a much larger problem.

The intervention was carefully calibrated because it shifted attention from the apparent cost line to the real one. Three million additional capacity to be able to turn the care budget is a lever no reorganisation can match. The three million did not go to bureaucracy but to the people who could deliver the top-100 work, the data system that held the chain together, and the coaching that would prepare line managers for the work after my departure. Without that prior investment, none of the other interventions would have been sustainable. It is also an Aiki manoeuvre¹ toward council and executive: rather than fighting the charge of too many officers, inverting the logic toward the actual lever. Whoever has to manage the purse should also be given the instruments to do so.

Self-indication. The second move was structurally heavier. Ordering and evaluating returned to one hand. The municipality issuing the indication is the same municipality that bears and funds the outcome. Indication was placed with a team of SKJ-registered professionals² led by an experienced team manager with practice background in youth care. SKJ registration means professional accountability tested by the professional body. That anchors substantive authority outside the hierarchical line. The team manager from practice provides the second element absent in most municipal access teams: authoritative substantive engagement with indication decisions, not as retrospective oversight but as weight in the primary process.

Signal and indication separated. The referral behaviour of GPs and certified institutions is, in most municipalities, the boundary of the steering space. Both are statutorily free referrers, and in practice two-thirds of youth-care demand bypasses the municipal access route entirely. In this municipality an arrangement was developed that separated the signalling function from the indication function. The GP could still indicate that a family needed youth care, and that signal was taken seriously. What the GP no longer decided alone was which youth care, at what intensity, for what duration, with which provider. My consistent position was that the municipality, as holder of the purse, decides what is deployed, when, and at what cost. Not contesting that youth care was needed, but weighing what was appropriate against what it cost.

What that separation does is respect the different competences that otherwise pile up at a single point. The GP knows the medical reality of the family. The access team knows the local provision, the budget, and the trade-off between lighter and heavier interventions. The provider knows their own intervention. By no longer collapsing these three into a single referral moment, no one has to judge what they are not positioned to judge. The Reach Bill³ currently before parliament works out precisely this separation at national level. What this municipality anticipated is becoming the standard others are now moving toward.

The top-100 as fractal approach. The top-100 was not a procurement decision concerning a hundred cases. It was the operational recognition that the scale of intervention itself had to be workable on multiple levels at once, because the pattern recurred at every level. On the street, in the neighbourhood, in the municipality, and even within the top-100 itself. Eighty per cent of the budget was not in twenty per cent of the families, as a Pareto assumption would predict. It was in ten per cent of the families, within which a fraction again carried the largest share. That undermines the most common objection to a top-100, namely that it operates selectively or stigmatises an arbitrarily delimited group. With a fractal distribution, the threshold is not arbitrary but instrumental: it is calibrated to what the delivery side can carry. The next round, on the top ten within the top hundred, does the same work at a different scale, with different intensity and different governance. It is a fractal answer to a fractal problem, and it broke with the conventional case-segmentation logic that municipalities tend to default to.

The top-100 was, moreover, designed as a continuous operation. Not a project with an end date, but a working process. When the first hundred had been professionally completed, the team moved on to the next. This eliminated the handover moment at which most transformation efforts revert to their old logic. Embedding was not a final ritual; it was the cycle itself.

The satisfaction test. What sharpened the instrumentation was the test applied across all cases. Not “has the care been scaled down”. Not “has the volume returned to baseline”. But “is this family satisfied with the approach, and is the professional satisfied as well”. That dual satisfaction blocked manipulation from both sides. A family content with too much support would not flag that itself, but the professional would. A professional content with insufficient support would be corrected by the family. The test thus shifted from the volume register to the effect register, and made it politically possible to leave volumes where appropriate, and move them where necessary. The financial outcome became a consequence of that calibration, not its goal.

The fractal data warehouse. A typical youth-care dashboard reports aggregated monthly figures to executive and council. That is precisely what dissociation feeds on: figures without faces, frozen into time series that no one can connect to a family any more. In this municipality a data system was built in which both the macro-aggregation and the seven families accounting for half the rise remained visible. This is not a reporting instrument but a diagnostic nervous system. It holds the chain together. It denies any link the option of saying “we are doing our task well” without being able to show which families that touches and what is happening to them. This is what oversight literature calls “direct source access”, here applied internally rather than as an external corrective.

Storytelling as a parallel accountability register. What no monitor can do, the stories did. Councillors were invited into rooms where they could hear how an ambulant practitioner actually works in a family carrying multi-generational trauma. Front-line workers spoke in their own words about what their work involves, by invitation. The position I have held throughout is that much of what we measure does not measure what is happening, and much of what is happening is not measured. This is not a critique of measurement. It is recognition that in a sector without objectifiable production, measurement must always be supplemented dialogically. A councillor who has once heard what ambulant work actually is no longer reads the next spreadsheet as if the figures were the reality. That is a lasting shift in how the political conversation can take place.

Aiki toward the council. The portfolio holder faced the conventional charge in council that the municipality was sitting in the professional’s chair. The conventional defence reads “we respect the professional”. That is exactly the Aiki that does not work, because it honours the attack by accepting that officer and professional are two different categories. The defence I gave the portfolio holder was sharper: the municipality is also a professional, and the disdain toward officers must stop. The attack was thus not answered but redirected toward its own premise. Aiki with moral grounding works only because it is true. A municipality that indicates at client level with SKJ-registered staff under an experienced practice-trained manager is, literally, professional delivery. Dismantling the symbolic opposition between “professional” and “officer” was the place where substance could land again.

The core team with the educated controller. The authority foundation rested on a small staff around the directorate position: strategic policy generic, strategic policy youth, controller, data. The controller had been educated in what the work actually was, and that detail rarely gets the attention it deserves in administrative descriptions. The conventional assumption is that the controller stands above the work numerically and tests it. In practice that only works when the controller genuinely grasps the logic of the work. Otherwise you get the familiar dynamic in which the controller reduces substance to what their instruments can hold. By educating the controller, a second substantive position emerged alongside the line manager — not a cost watcher but a twin half of the delivery. As a result, financial and substantive logic pointed in the same direction, where in most municipalities they sit at right angles to each other.

The scenario study as exercise. In the first half of 2019 I proposed a scenario study. Not to arrive at a choice between four futures, but to take management and policy through the experience of thinking in terms of navigation rather than planning. A scenario study is a serviceable didactic tool for that. Anyone working through the exercise of plotting four different futures on an axis cross and locating stakeholders within them experiences, somewhere along the way, that reality will not coincide with any of the four quadrants. That is precisely the insight conventional policy practice lacks, and it was needed here so that the delivery phase could be lived as a course rather than a plan.

In the second round we reduced the four scenarios to two: Playmobil and Lego⁴. The advice to council read “Lego, without structural intervention”. Honestly: I had reached that conclusion for myself at the outset of the trajectory. What the scenario study did was not leave the choice open but walk the route toward it with the people who would have to carry it. That distinction is, in my experience, the difference between a scenario study that works and one that hangs as legitimation above an already-taken decision. The conventional outcome of a scenario study into the configuration of the social domain is a reorganisation. By arguing within the scenario logic that the existing structure offered room provided that roles were worked with and behaviour intervened upon, space was created for substance. An organisation not occupied with reconfiguring itself retains capacity for delivery.

The advice was tied to a transformation-manager assignment with an explicit end date: Q3 2019 to Q2 2020. “Departure of the transformation manager” was the final line in the work plan. That is not a footnote. It is the central design choice. An intervention that should not become dependent on its executor must build in its own departure from the start. The measures around coaching line managers, developing career paths for staff, and management development for the social domain are the construction by which, after departure, the work could reproduce itself. Embedding as primary test, not added at the end but designed into the architecture.

The pattern beneath

When these eleven interventions are placed alongside each other, a pattern emerges that formed gradually rather than being designed in advance. It concerns the simultaneous lifting of dissociations at different levels of the chain. Between indicating and funding, by self-indicating. Between apparent and actual cost line, by over-committing on the saving in exchange for additional capacity. Between signalling and indicating, by respecting the GP’s medical signal function without burdening it with the indication authority. Between ordering and evaluating, through the top-100 method with SKJ-registered delivery. Between pattern and face, through the fractal data warehouse. Between measurement and reality, through storytelling as parallel accountability register. Between municipality and profession, through symbolic dismantling of the disdain. Between project and line, through the continuous top-100 cycle. Between interim work and organisation, through the transformation manager with explicit end date. And, not least, between scenario thinking and acting, by using the scenario study to support a recommendation that contradicted its conventional outcome.

None of these moves was a magic bullet on its own. Together they formed what earlier Statecraft work calls a constructed setting, a connection in which substance could land again, except that here it was not one setting but eight, mutually reinforcing. The financial saving was the measurable indicator of a far less measurable shift. The chain reattached to itself.

What this illustrates at the level of interim work is that the most robust change is the one in which there is nothing left to hand over, because the work reproduces itself. The Interim Cycle developed in earlier work treats the handover phase as the most difficult. In this case the handover phase was solved by abolishing it. The cycle itself became the work. Anyone who designs a top-100 as a continuous process is no longer designing a project. Anyone who ends a transformation-manager assignment on a date already in the work plan is designing their own redundancy. That is a design principle that lends itself well to formulation as the Aiki method in its institutional variant: redirect the energy that normally flows toward resistance into redesign. Not by working against the organisation, but by formulating, within its own logic, the choices that free its delivery.

The limits

It would be misleading to present this case as a generally applicable precedent. Three elements could be organised at municipal scale that are structurally absent at system level.

The legal space was there. Since 2015, Dutch municipalities have had the authority to indicate themselves, to select partnerships and to govern access. In many municipalities that authority has been delegated away to youth regions and providers. In this municipality it was actively reclaimed. That was not a structural turning point; it was a use of authority that other municipalities could equally choose.

The scale was workable. A municipality of just over a hundred thousand inhabitants is large enough for professional capacity and small enough for families to be known. A top-100 is workable at that scale. In a major metropolitan municipality, or in a youth region of six hundred thousand inhabitants, the same method breaks under its own weight. The fractality remains, but the authority to act fractally dilutes across too many actors.

The administrative coherence was uncommon. One portfolio holder, one council, one directorate, one vision, with a cluster director given the authority to carry it substantively. At system level that coherence is unthinkable. Council, executive, region, ministry, professional body, inspectorate and provider field do not constitute a space where everyone can be brought to the table to give substance its weight together.

There is one further substantive limit honesty requires acknowledging. With providers it remained difficult. The partnership principle stayed caught in the practice of open-house procurement, and the shift from heavy to light care ran up against national contracting logics and the lobby of incumbent providers. The 2019 audit-office report rightly noted this.

Three years after my departure the chief executive officer rang me to report that the targets had been achieved and the youth-care finances were back in balance. I had not expected that call. The conventional experience with transformation projects is that they post results during the work and revert to old patterns within three years afterwards. Here the architecture held without the executor who had set it up. That is not a definitive answer to the embedding question. Whether the centrifugal pressure of the youth region, the national reform agenda and the supra-regional arrangements eventually pulls the system back into its dissociative mode is the open question from the Dissociated Organisations paper: when the system itself is in motion, and the patterns we recognise as neglect are present everywhere in comparable measure, what is the reference point against which recovery can be calibrated? But the people who carried the work after my departure were evidently willing to hold the architecture. That is more than I had any right to expect.

What this says

For the public administration of the social domain, three observations from this trajectory became clearer to me than I had found them in the literature. The first is that the mathematical structure of the problem must determine the instrument. A fractal distribution requires fractal intervention logic. Programmes designed around averages structurally miss their target. They produce volume, not effect.

The second is that change without reorganisation can be faster and more durable than the conventional reflex assumes. An organisation not occupied with reconfiguring itself retains capacity for delivery. The cyclical self-distraction of Dutch public administration, in which a new administrative layer is erected every decade, is breakable at municipal level. It requires the courage to argue, within the scenario logic, that the structure does not need to change.

The third is that connection is not a therapeutic metaphor but a design principle. Settings in which substance carries more weight than procedure are not naturally present in an institutional architecture, but they are constructible. The opening conference, the satisfaction test, the fractal data warehouse, the storytelling sessions, the core team with the educated controller: these are all variants of the same constructed setting. Their absence is what causes so many good intentions in the social domain to fail. Their presence is what makes recovery possible — not as a magic bullet but as a design choice that has to be made, again and again.

Anyone now, six years on, taking up a comparable position can read this case as method or as precedent. Neither does it justice. What it offers is an example of what was achievable in a specific configuration, with authority, scale, core team and political arrangement aligned, and with an executor who attempted to build in his own departure from the start. The deeper question of whether this was retraining toward a shifted norm or genuine recovery remains open. The measurable part was achieved. Whether the architecture also structurally broke the pattern of dissociation, or merely displaced it temporarily, lies beyond what I can judge from a distance. The work undertaken in those years has earned the question being asked.

Footnotes

¹ The Aiki method, developed in earlier work, draws on the martial-arts principle of redirecting force rather than meeting it head-on. In administrative practice it means refusing to engage on the terrain the opponent has chosen, instead reorienting the encounter onto an axis where the underlying truth becomes the load-bearing structure. It only works when the redirection is honest; it is not a rhetorical trick.

² SKJ (Stichting Kwaliteitsregister Jeugd) is the Dutch register for youth-care professionals. Its function is comparable to that of the Health and Care Professions Council in the United Kingdom. Registration confers professional accountability that operates independently of the employing organisation.

³ Wetsvoorstel Reikwijdte jeugdhulpplicht — a Dutch government bill clarifying the boundaries of municipal duty in youth care, including the relationship between medical signalling and municipal indication. It is at the time of writing in the parliamentary process.

⁴ The contrast was meant illustratively. Playmobil = continue developing the existing configuration, fixed pieces with predetermined functions. Lego = reorganise around target groups rather than functional domains, with reusable building blocks. The framing made the abstract scenario language concrete enough for officers, councillors and providers to engage with.

Colophon

“Connection as Design Principle” is a Statecraft publication, published as a case study alongside the closed series Dissociated Organisations (April–May 2026). It works out, in one concrete assignment, what the series formulated as a repertoire of action. The forthcoming book The Direction of Movement: Interim Management in the Public Sector (manuscript in preparation) will include this case in the chapter on the handover phase of the Interim Cycle, because it shows that the most robust handover is the one in which there is nothing left to hand over.

Responses and counter-arguments via Statecraft.nl.

Jacob Huibers is an interim manager with more than twenty years of experience in the Dutch public sector. He has worked as cluster manager, cluster director, and quartermaster for municipalities ranging from fifty thousand to over two hundred thousand inhabitants, and for inter-municipal collaborative bodies. Statecraft is his platform for strategic reflection on public-sector execution.