When rational choices leave less room to act
The election results in Saxony-Anhalt created a political stalemate that seems illogical at first glance. The largest party, the AfD, won nearly 44 percent of the vote but was unable to form a government. Virtually all other parties ruled out cooperation from the outset, which greatly reduced the scope for forming a workable government. How can a democratic process produce such an ungovernable result?
This creates a curious problem. Political parties may have perfectly understandable reasons for the positions they take. One party may conclude that cooperating with another party is unacceptable. Another may decide that compromises would undermine the position on which it was elected. Still other parties may seek unlikely coalitions because they consider the alternatives to be worse.
Look at each decision individually, and it may make sense. But as soon as those choices begin to interact with one another, the collective scope for action may actually shrink. That’s exactly where alignment becomes interesting.
Alignment is not the same as agreement
Alignment is easily confused with consensus: getting everyone on the same page, resolving differences, and reaching agreement on a common direction. That cannot be the norm in a democracy. Political disagreement is not a flaw in the system, but part of the system.
The more useful question is whether those differences can still be translated into legitimate decisions and effective governance. This suggests an important distinction.
We can distinguish between two types of alignment. The first is substantive alignment: do the actors agree on the destination? Do they share the same goals or ambitions? The second—and, for a robust system, far more important—is systemic alignment: do the actors agree on the traffic rules? Even if they want to go to different destinations, can they still function within the relationships, incentives, and rules that keep the system workable?
A democracy can function perfectly well without everyone wanting the same thing, but not without sufficient agreement on the rules and institutions within which differences are accommodated.
The Alignment Paradox
This brings us to a paradox that extends far beyond politics. Actors can make choices that are entirely rational from their own perspective, while the interaction between those choices gradually narrows the scope for collective action.
One decision changes the possibilities for others. They adapt, and their response, in turn, changes the possibilities once again. Positions become harder to change, and the number of viable combinations decreases. No one needs to behave irrationally to cause the system to break down, and it is precisely that distinction that matters.
If irrational behavior is the problem, changing that behavior may be enough. But if people are behaving rationally within the system in which they operate, telling them to cooperate better will likely do little to solve the problem. In that case, the system itself may need attention.
Relationships are part of the architecture
Relationships are often seen as the softer side of organizations: trust, communication, stakeholder engagement, chemistry. But relationships also serve a structural function. They determine who can speak with whom, negotiate, exchange information, make arrangements, and ultimately make joint decisions.
When an important direct relationship is no longer available, part of that connecting structure disappears. The instinctive reaction is often to restore the relationship: bring people together, facilitate a conversation, and look for common ground.
Sometimes that works, but sometimes the differences are real. In that case, the more useful question is not how we can still reach an agreement, but whether the relational architecture itself needs to change.
For example, a mediator can bridge gaps between positions that cannot work together directly. Representation can change, decision-making can be organized differently, and a different combination of actors can create a path forward. The goal is not necessarily to get people to agree, but to make differences workable.
I do not claim to know what the correct approach is within the specific constitutional and political context of Saxony-Anhalt. The case raises the question; it does not provide us with the answer. But it makes the mechanism exceptionally clear.
Why Rational Actors Get Stuck
This is where various systemic dimensions begin to interact.
The *relationships* determine which actors can collaborate with one another. The *incentives* explain why maintaining or rejecting those relationships might make sense. And the *organizational structure* determines what decisions can still be made when certain relationships are lost.
These dimensions can then reinforce one another. Suppose a particular relationship becomes costly. The incentive to avoid it increases. This reduces organizational flexibility, after which the remaining options can in turn make it even more costly to change course.
What began as a reasonable choice thus becomes part of a feedback loop. That loop is ultimately more important than any individual score within a framework.
Alignment is therefore not simply a characteristic of the actors involved. It arises from the way their choices interact with one another.
A much more everyday example
Imagine a complex program involving a ministry, regional governments, companies, and researchers. The ministry wants assurance before committing more funds. A company wants commercial certainty before investing. A regional government wants clarity on local benefits before providing political support. Researchers do not want to guarantee results before the evidence exists. The program manager cannot make decisions that formally fall under the partners’ purview.
No one does anything absurd. And yet, nothing changes.
It’s tempting to call this a collaboration problem and organize yet another meeting. But perhaps collaboration isn’t the problem. The system may be requiring each actor to make a decision that’s difficult to justify as long as someone else doesn’t make the first move.
If that is the case, another discussion about the shared ambition will not necessarily help. Decision-making authority may need to change. Commitments may need to be phased in differently. Incentives need to be adjusted. Ownership needs to be clarified. Or a different relationship between the stakeholders may be necessary.
The intervention depends on the mechanism.
Before we ask for more cooperation
The first step is an accurate diagnosis. Instead of focusing on the willingness to cooperate, a systemic analysis asks three key questions:
1. Is the behavior of the actors unreasonable, or is it a logical response to the system in which they operate?
2. What specific relationships, incentives, or rules are perpetuating the current impasse?
3. What changes to those system characteristics are needed to make a decision possible, despite differing interests?
These questions are deliberately not based on consensus. Sometimes more agreement is exactly what's needed. Sometimes it isn't.
Designing for Diversity
Any complex change inevitably brings together actors with different interests and incentives. A governance model that works well only as long as there is consensus is therefore not very robust.
The real challenge is to design arrangements that function properly precisely when interests diverge. That is where the distinction between substantive and systemic alignment becomes critical.
Substantive alignment requires: Do we want the same thing?
Systemic alignment asks: Can we still get things done when that’s not the case?
That second question has very practical implications. If the answer is no, friction increases, the decision-making slows down, risks pile up, and results suffer.
Alignment, therefore, is not a preference for harmony. It is about organizing differences well enough to be able to act.
Saxony-Anhalt is an extreme and politically charged example. The scale and the stakes are unique, but the underlying mechanism is not. Precisely because the conflicts and limitations are so clearly visible there, this case study reveals dynamics that often remain hidden in the everyday context of organizations and programs.
The true test of alignment is not whether actors agree with one another when reaching consensus is easy. It is whether their differences can still be managed well enough to enable them to act when they do not agree with one another.