You've probably heard politicians promise that agencies will "work together" to solve complex problems. Homelessness? Get housing, health, and employment agencies collaborating. Drug addiction? Bring together law enforcement, healthcare, and social services. It sounds so sensible. So why does it almost never work the way anyone imagines?

The gap between coordination on paper and coordination in practice is one of the great unspoken realities of government. Agencies that are supposed to collaborate often end up competing, hiding information, and pointing fingers when things go wrong. Understanding why this happens isn't about cynicism—it's about designing policies that might actually succeed.

Here's something they don't teach in civics class: every government agency is, at its core, a survival machine. It needs budget, staff, and political support to keep existing. And when another agency shows up wanting to "partner," the first instinct isn't collaboration—it's suspicion. What do they really want? Are they trying to take our funding? Our mission? Our staff?

This isn't paranoia. It's learned behavior. Agencies that have watched partners absorb their programs, claim credit for joint successes, or blame them for shared failures develop institutional antibodies. The mid-level managers who remember these betrayals become the senior leaders who quietly sabotage the next big coordination initiative. They'll attend the meetings, sign the memoranda of understanding, and then do absolutely nothing that threatens their agency's autonomy.

The tragedy is that this self-protection is often rational. In a world of shrinking budgets and political uncertainty, agencies that don't guard their territory often lose it. The incentive structure rewards building walls, not bridges. So when a mayor or governor announces a bold new interagency task force, the smart money bets on polite cooperation in public and quiet resistance in private.

Takeaway

Agencies aren't being selfish when they protect turf—they're responding rationally to incentive structures that punish vulnerability and reward self-sufficiency.

If coordination is the goal, information sharing is the currency. Agencies solving the same problem—say, helping a homeless veteran find housing—each hold pieces of the puzzle. The VA knows the medical history. The housing authority knows available units. The employment office knows job training options. Put it together, and you could actually help someone. Keep it apart, and you get what we usually get: the same person telling their story to seven different intake workers.

So why don't agencies share? Start with the technical barriers: different computer systems that don't talk to each other, data formats that can't be translated, privacy regulations that lawyers interpret as "share nothing, ever." These are real obstacles. But they're also convenient excuses. Agencies often could solve the technical problems if they wanted to. They frequently don't want to.

Information is power. The agency that holds unique data has leverage—in budget negotiations, in policy debates, in the daily jockeying for influence. Sharing that data means sharing that power. It also means exposure: if another agency can see your outcomes, they can criticize your performance. Data sharing agreements get signed with great fanfare. The actual sharing happens quietly, reluctantly, and incompletely—if it happens at all.

Takeaway

Data sharing fails not because of technology but because information represents power, and agencies rarely surrender power voluntarily.

Here's the cruelest irony of interagency coordination: the more agencies you involve, the less accountability you create. When one agency runs a program, failure has an address. Someone gets blamed, investigated, maybe fired. When five agencies share responsibility? Failure becomes an orphan that nobody claims.

Politicians love announcing partnerships precisely because they diffuse blame. If the homeless initiative fails, was it the housing authority? Health services? The workforce development board? The mayor's office that was supposed to coordinate? Everyone can point at everyone else. Post-mortems become finger-pointing sessions where each agency explains why their part worked fine—the problem was obviously the other guys.

This accountability vacuum affects behavior before anything even goes wrong. When staff know that failure can be blamed on partners, the incentive for excellence diminishes. Why work nights and weekends on a project when you'll share credit for success but can dodge blame for failure? The result is what researchers call "satisficing"—doing just enough to avoid being the obvious weak link, but never truly owning the outcome. Shared responsibility becomes, in practice, shared mediocrity.

Takeaway

Collaboration that divides responsibility among many agencies often means no one truly owns the outcome—and what no one owns, no one fights for.

None of this means coordination is impossible—just that it's far harder than the announcement press conferences suggest. The initiatives that actually work usually have someone with real authority who can bang heads together, shared performance metrics that matter to everyone's budget, and enough time to build trust through small successes before tackling big ones.

For citizens watching government promise collaboration, a healthy skepticism is warranted. The question isn't whether agencies say they'll work together. It's whether anyone has changed the incentives that make working apart the safer choice.