Lean startup nonsense

The "lean startup" model could indeed help government run better. Under lean startup, a business (or government) creates small "minimum viable products" and tests them with consumers.

The object is to test the hypothesis that consumers will enjoy/purchase X product. Lean startups want to iterate, release and test products quickly, all the while using the knowledge gained from testing user behavior to make a better product. Basically you're testing to find out what consumers want (and why they want it) and quickly drop whatever doesn't work for the consumer. The result is a successful business that produces products people want at a lower cost.

Sounds great and the concept of applying the lean startup method to government has Pando Daily writers and Eric Ries (the guy who coined the term lean startup) excited about the prospects.

An article by Hamish McKenzie in Pando Daily provides minimal detail on how government is using lean startup, but basically a former executive level administrator at Health and Human Services is pairing entrepreneurs with government bureaucrats to produce "results" in six months.

McKenzie writes,

The White House was applying Lean Startup concepts in the Innovation Fellows program, which pairs top entrepreneurs from the private sector with top innovators in government to collaborate on key tasks that aim to deliver significant results in six months. The teams work to build products that address needs such as making government data more useful and accessible to the public, improving awareness around public health records, and personalizing the government’s online services.

[Todd] Park was effusive about the merits of the Lean Startup way. “The government in and of itself isn’t a startup, but change instincts inside the government absolutely are,” he said. “And Lean Startup works.” He described it as a “perfect fit” with how the government has to change.

Forgive me for being skeptical, but I don't see the lean startup model working in a virtual monopoly where politicians and bureaucrats have no skin in the game (nothing to lose by spending other people's money). There are also associated problems with concentrated interests and dispersed costs (government beneficiaries - read special interest groups - are highly vocal, taxpayers less so). This issue with concentrated interests results in decisions on policy prescriptions being political, not based on science. Finally, government has a problem with how it earns revenue. You see (Eric, Hamish, Micheal, Todd et. al) government earns revenue regardless of whether the programs are successful or not. Government has the power to tax and to take what it needs whether the services are good or bad (little to no incentive to do good).

Examples abound, but I have one for you that is rich with irony.

Todd Park (government CTO quoted above) worked for Health and Human Services, the same department which suppressed a study for 3 years on the Head Start program. The study found no long term statistical impact on students in the program. This despite spending over $150 billion on the program since its inception 40 years ago.

Head Start still exists and the government still spends billions on the program. If that is lean startup, I will eat my underpants.

If that wasn't bad enough, during the same time HHS was suppressing a study on how they wasted billions producing no results, the Democrat controlled congress voted to eliminate the DC Opportunity Scholarship program which, at worst, produced a 20 point improvement in graduation rates among low-income students in DC.

Lean startups don't spend billions on projects that don't work and cancel projects that do. Lean startups also don't have the ability to take, by force, revenue from their customers...


Consumers of the scholarship program wanted it (see the video above), but the teacher union (a powerful and influential special interest group) feared the program threatened union power and control over education. The union pushed Democrats to kill the scholarship program despite overwhelming evidence that school choice programs work.

The fundamentals of government (how government ACTUALLY operates) doesn't allow for the lean startup method. Its a nice thought, but it just doesn't work out that way right now... For now, we'll get platitudes from government bureaucrats and the journalists who believe, without question, everything they say.