New Public Management and the Contract State: The model that was never fit for purpose

29/7/2020

Every few years, someone revisits the public sector reforms in Aotearoa and declares them a success story. They point to contractualism, outputs, ministerial purchasing power, purchaser-provider splits and accrual accounting. Maybe they even quote Allen Schick’s 1998 piece, “Why Most Developing Countries Should Not Try New Zealand’s Reforms,” waving it around like an endorsement.

It’s not. Not even close.

Schick’s warning was clear: don’t try to graft this system onto countries with immature economies, weak enforcement, or politicised civil services. His view was that the New Zealand reforms could only function, if at all, within a very narrow set of conditions: strong rule of law, reliable and steady markets, a culture of learning and compliance, strong regulatory cultures and a willingness to invest in capability to keep it relevant. Without those, the model not only falls short, but it becomes destructive.

But here’s the real kicker: those preconditions no longer hold true here, if they ever did.

This post continues the PIF Series argument: that the idea of practical efficiency in public sector reform, particularly as imagined in the New Zealand model, was never achievable. Not because people failed to implement it well, but because the model itself was flawed due to conceptual mistakes.

Let’s be blunt: government can’t contract with itself and expect to behave like a market. The New Zealand model was obsessed with creating quasi-markets. Outputs were priced, departments were treated as providers, and contracts were everywhere. But markets rely on real exit. If your contractor fails, you find another. Governments don’t have that luxury. When a department underperforms, there’s nowhere else to go. When there’s a “market,” how are you going to conjure up a health provider for Fiordland? The whole contractual logic collapses. Better service mapping would have been the correct solution. Instead, we got economic rationalism that completely misunderstood delivery in the real world.

But there’s a more profound truth here.

Efficiency wasn’t just undone by structure: it was also undone by underinvestment.

The promise of efficiency and effectiveness only holds if you’re willing to build the foundations. This entails making substantial investments in technology, data infrastructure, and capabilities. Without those, performance data can’t flow, learning can’t occur, and the information asymmetries that the reforms claimed to solve remain just as severe, if not worse.

The PIF system-level data backs this up. Only a third of the assessed output areas had efficiency measures in place. Let me reiterate: just over 30% of output areas, across more than 70 reviews, had meaningful efficiency measures in place. The same was true for effectiveness: just over 50% of output areas had effectiveness measures in place.

The chronic underinvestment in digital systems, legacy platforms, disconnected reporting, and minimal feedback loops led to the failure of new public management reforms. Without the tech backbone, the reform logic broke. And without a learning culture, which new public managers refused to endorse, even the most perfect data rotted on the shelf.

So what are we left with? A model chasing efficiency, built on contract structures with no real competition, operating inside systems with no visibility, and no serious tools for learning. The result? Poor efficiency ratings, demotivated staff, and systems still struggling to deliver on purpose, let alone outputs. That is the legacy of new public management in Aotearoa.

The reforms narrowed public service to a counting exercise. Outputs replaced purpose. Performance agreements replaced public trust. Ticking boxes, not making change, became the measure of success.

In doing so, we trained a generation of public managers to think in contracts and spreadsheets, rather than judgment and stewardship. We embedded a view that if something wasn’t specified in an agreement, it wasn’t their problem. The consequences have been obvious: fragmented accountability, perverse incentives, and a complete mismatch between what’s being delivered and what communities need.

This is the world the PIF Series keeps returning to: one where new public management promised efficiency but delivered constraint and underdelivery—very busy milk monitors, but little else.

Schick’s critique focused on developing countries, but his insights remain applicable to most nation-states. We all act as if our systems are rule-based and high-trust. Still, much of what happens runs on informal mechanisms: relationships, officials going above and beyond, backchannels, tacit and local knowledge, and moral and ethical judgment. We don’t call it informality. We call it political nous or system sense.

The risk comes when you replace those informal working practices with brittle contracts, then pretend you’ve strengthened the system. With new public management, the opposite happened: we deskilled it.

If Schick were revisiting his 1998 paper today, I suspect he’d say the model has run its course. It was built on false equivalences between government and the business sector. Contractualism gave us the illusion of control while disabling adaptive capability. Chasing “efficiency” through faux markets has made the system slower, more expensive, and less capable of navigating complexity.

But I lay the accountability at the feet of the new public management apostles. Efficiency, at least as it was imagined, was never truly possible. The institutional context wouldn’t allow it. The political environment wouldn’t support it. The political class lacked the courage to demand efficiency from all domains, let alone invest in pursuing efficiencies. And the very model used to chase it actively undermined the conditions needed for good government.

The reforms of the 1988 state sector didn’t fail because they weren’t applied properly.

They failed because their approach was conceptually flawed.

Time to move on.