How Discredits Liberalism

The Internet does weird things to people, including to some of my friends. One of those weirdnesses comes from my tech-savvy friends on the left, who sincerely embrace an agile, entrepreneurial, bottom-up culture in their professional and voluntary pursuits, yet forcefully argue for the top-down paternalism of forcing people to buy health insurance, imposed by a bureaucracy that can’t build a website. 

Does the disastrous launch of automatically mean that the same bureaucracy won’t eventually get it right on Obamacare more broadly? Of course not. But the odds aren’t good, for reasons that go far beyond the website itself. is an object lesson in the perils of big bureaucracy and crony capitalism. These issues are at the heart of the broader conservative critique of big government. Using an earlier tweet of mine, Matt Yglesias tries to argue that doesn’t discredit liberalism. He is right to be concerned that it might. 

No one has a better understanding of this issue than Clay Johnson, a friend who traded partisan politics (including helping to elect the president in the first place) for helping government build better technology. He is furious, and rightly points out that the issue stems from federal procurement laws.

But things like bad procurement laws don’t come out of nowhere. They arise in dysfunctional cultures that are the result of highly politicized, CYA-driven decisionmaking processes. Part of the blame lies with successive Congresses. But part of it, if we’re being honest, is the rigidly bureaucratic culture of government itself. No one who values getting things done quickly and efficiently would ever build a system like the one we have now. 

Congress could design and federal IT regulators could enforce a new procurement law that was the epitome of agility and nimbleness, one that would allow for tiger teams of Mark Zuckerberg and Sergey Brin’s best engineers to tackle things with “Move Fast and Break Things” gusto. But such a law would be the antithesis of the modern administrative state that is a product of 20th century liberalism. 

I get that progressive techies didn’t set things up that way, and left to their own devices, can produce pretty good results (see the 2012 Obama campaign). But the governing philosophy they support is at odds with the startup mentality they also embrace. For better or worse, the impulse for a bigger government that makes more health care decisions goes hand in hand with bad procurement laws administered by a unionized federal workforce bullish on routine and bearish on innovation. 

This is how we get a bureaucracy that, when it comes to technology, moves slow and still breaks things. 

Nor is this critique purely a partisan one. In 2005, I worked in the CIO’s office at a federal agency, and saw how the big IT contractors (like CGI) served as an extension of the government itself, colluding with the federal bureaucracy to shut out upstart competitors. At the time, the Bush Administration was pushing to outsource more of the government’s technology work. Ostensibly, private contractors would compete with the federal workforce for big chunks of the IT work. In our case, though, the incumbent contractor found a way to join forces with the feds. Knowing that the fix was in, no one else would bid. The incumbent public-private consortium was awarded a $90 million-a-year contract, effectively in a no-bid process, all done under the aegis of the Bush Administration’s push for more private sector involvement. 

The Obama Administration came into office with some good impulses, even talking up procurement reform. But it has utterly failed to deliver on those promises. 

We can and should talk about the specifics of procurement reform. Fixing a badly broken federal technology infrastructure isn’t, and shouldn’t be, a partisan issue. 

But this does not allow us to ignore the broader question of what types of organizing structures tend to produce the best results in actually delivering things that work: private, for-profit technology companies serving rational, profit-maximizing consumers (like Apple or Google), nominally private companies that ultimately serve their bureaucratic masters (like CGI), or the government itself. 

After, the answer is clearer than ever before. And so are the public policy implications of that answer. 

  1. wanorris reblogged this from patrickruffini and added:
    The usual cited figure is that 50% of all IT projects fail. This figure is not specific to the public sector. So it’s no...
  2. caffeinated-zone reblogged this from patrickruffini
  3. patrickruffini posted this