Stimulus as Software Project
The whole economic stimulus package thing just feels wrong to me. And it’s not the whole debate over “spending” vs “tax cuts”, blah blah. It feels wrong to me because it reminds me of some huge software project with a massive Gantt chart that’s supposed to end in two years. And in my gut, I know that software projects like that fail. Often in a big spectacular way, but if not then in thousands of small ways by building systems that aren’t what users need, cost too much, run late, whatever.
Everything I know about software development tells me that what we need is agile stimulus. Pick 100 things, and try them. Assess the results, double the ones that work, drop the ones that don’t. Repeat. If I remember right, FDR tried a zillion different things until something worked.
I know the economy is hosed and all, but this whole plan just feels like a crazy expensive project doomed to failure. Am I crazy?
Hi! My name is Alex Miller and I live in St. Louis. I write code for a living and currently work for 
That’s hope and change we can’t afford!
We’re screwed in 09!
A more agile/bottom-up based solution will still be needed, as I understand the stimulus packages are mostly there to avoid a total meltdown as in 29′. As someone who potentially saw himself included in a down-sizing round recently, it felt only too real how we first and foremost need to fix the original credit crisis of 2007 – put some trust back in the marked and stop the vicious cycle of banks clinging on to whatever little they have.
Or to use an analogy. While pumping air into flat tires might not by itself make a vehicle run, it will keep it running longer on an existing momentum. The hard thing I guess is to find out and exclude vehicles whose low momentum have deemed them doomed anyway.
As an individual with comfortable work insurance, solid education and living in a social welfare state, I would be ok no matter what. It’s when looking around me at the waitress with two kids to feed, Walmart employee who can not afford a doctor etc. that I get worried. So let me ask you this, if your country did not already owe a gazillion $ to China, would you feel any different about such a stimulus plan? Or does it boil down to your trust in the free marked i.e. not rewarding incapable businesses and managers?
I had been thinking the same thing.
I have been listening to the EconTalk podcast and they have made similar points numerous times, but then Russ Roberts in interested in emergent behaviour. He recently interviewed Eric Raymond.
When he talked with William Easterly about foreign aid, they talked about succeeding from the bottom up, not the top down and not imposing systems but letting the systems emerge from the bottom.
All along I had been thing about how well it conformed to Agile projects.
There must be 100 things to try in the stimulus bill, and unlike a big software project, not a whole lot of interdependence between them. Asking Congress to fund them all at once is a lot to ask, but it does get the ball rolling quickly. The funding can be changed later.
Maybe you’re looking at it wrong: if the purpose of such a plan was to achieve a long list of specific goals by a specific date, it would be doomed to failure. But the purpose is much simpler: to spend a bunch of money, quickly. The specific goals and projects are almost a sideshow.
Now as to whether it’s really a good idea, there is much debate. But as to whether our government can successfully print/loan/create a pile of money and spend it, there is little question.