Monday, September 13, 2010

Fixing the current mess

The Silicon Graybeard riffs on my post about regulatory-imposed stagnation, and quite a riff it is. The key issue he raises is, while it's all well and good to rant about the Fed.Gov's screw-ups, how do we fix things?
What can we do? We can't form a "skunkworks country" that can get around our laws and create a more mobile, productive society. We only have one option: we have to create a national process, like industries do, to become more "lean, mean and low to the ground". Get rid of superfluous laws. We simply must reduce the size of the CFR and reduce the destruction caused by the regulation and litigation in our society. To me, Tort Reform is absolutely essential. A big part of the industrial lean activities is to study what policies need to be gotten rid of because "we've always done it that way". The same should be done with the CFR.
Ah, but here's the rub: how do you know which laws are "superfluous"? We sure as heck can't trust the politicians or the agencies to tell us honestly. Assuming that they even know.

But the Graybeard did a great job dissecting the problem from an engineer's viewpoint. You learn a lot that way, a lot that the Fed.Gov could benefit from. You can also dissect the problem from a business management perspective; the Fed.Gov could benefit from this, too. Since I do this sort of thing for a living, I'm fixin' to build on Greybeard's work to construct a framework to fix at least part of this mess.

How do you know what's superfluous?

Business does Cost/Benefit analysis all the time. Businesses that do this poorly go out of business, those that do it well tend to prosper. Unsurprisingly, it's a well understood field. Graphically, you can plot proposals in a matrix - the more benefit, the higher up the proposal goes. The higher the cost, the further to the right the proposal goes. Businesses love proposals in the upper left corner (big payoff, easy to do) and hate ones in the lower right corner (not much payoff, but lots of work).


So when we think of the Code of Federal Regulation (CFR), how do we use this? We need measurements that make sense for the vertical and horizontal axes. Benefit is less straight forward, but we'll come back to that in a bit - if you don't quantify it, you can't measure it. Duh. The "cost" needs to be something that will have long-term traction with the voters. I propose "Jobs lost" as the appropriate measure. Our matrix is used like this:

The upper left quadrant shouldn't be touched. Pure libertarians will want to reduce all this because they think that axiomatically government should be reduced. Back in the Real World, programs that the public believes provide value for their tax dollar are pretty popular. It's hard to see any political coalition that could be put together to actually do anything here.

The upper right quadrant are where the big political battles of our day are fought. Whether these programs should exist or not, it's hard to see any sort of durable, long-term coalition making progress here.

The lower left quadrant is filled with trivialities. Sure, we don't get much for our tax money, but we're not losing much here. Ignore all this.

All our thought should be going into the lower right hand quadrant. Whatever's here is unjustifiable, and we can see even Democrats plausibly going for the jobs rather than for whatever the ostensible regulatory program. This is where it makes the most sense to start, because nobody will fight too hard for what's here.

How do you measure the cost?

Well, we can't ask the government agencies. Everything they do is mission critical, we're told. So we need to ask businesses. Fortunately, Al Gore just put up this big Intarwebz thingie. We can ask people. Lots of folks won't have any idea, but some folks - say, small business people - will get specific. That's where your raw material comes from.

Computing tools make analyzing this data child's play. Sort by number of complaints per regulation - a whole metric ton of regulations won't get ANY complaints: those are in the lower left "Don't Care" quadrant. A bunch of regulations will get a bunch of complaints, and the question then becomes OK we know the cost; what's the benefit?

Measuring the benefit

This also is easy. Each regulatory agency will have to tell us what the benefit is, and how they measured it. Put this up on the web, and people will rip the initial weak analysis to shreds. Send it back and make them do better, and iterate, and at this point we will start to see one of two possibilities:
  • Nobody can agree on benefit. Maybe this is in the upper right quadrant, but there won't be a stable political coalition to support getting rid of it. Move on to the next one.
  • There's no coherent benefit statement, meaning the benefits are unmeasurable. Treat it as lower right hand quadrant stuff, ripe for repeal. Give its supporters a chance to show the benefit - if they can't, we shouldn't keep it.
Then kill everything in the lower right quadrant. The beauty of this is that it's probably political death for anyone to stick up for these, as long as the electorate is in a foul mood. And this is where we get to the absolute key concept:

All of the lower right hand quadrant regulations benefit someone. Those someones will lobby to save them, and lobby hard. It's up to us as voters to punish anyone who puts these lobbyists ahead of us. We start this by voting everyone out, so they know we're serious. They'll be tempted by the lobbyists, and the campaign contributions. It's our job to make them think it's poisoned money.

And if the lobbyists win, and only 1% of the regulations actually get canned, the next election cycle will make this one look like a walk in the park for any incumbent. Once the bums are all thrown out, we do it again, harder. Like I said, this is the makings of a long term, durable coalition.

Ultimately, it's hard not to see 20% of the CFR as being anything other than lower right hand quadrant stuff. Superfluous. After one round eliminating most of that, you could probably go back for a second round that would get another 10%. Heck, maybe even more is a worthwhile target. There may be a dynamic that could grow out of this, requiring a justification for all new programs, but that's a topic for another day.

I think that there is a long term, durable political coalition that could be built around this idea. Using the Internet as a data input mechanism -- as a "Skunk Works", so to speak - engages the portion of society most impacted by stupid regulation, which builds an automatic support base. Focusing on improving government - rather than eliminating it - moves the coalition to the center, which is where all long term movements have to live in this country. Focusing on the practical, rather than the ideological, puts this in the sweet spot of the American Electorate. Fix what's broken, or eliminate it.

Thanks to the Graybeard for helping crystallize a bunch of ideas I've been trying to corral for a while. You really should RTWT.

6 comments:

roy in nipomo said...

Nice idea. Too bad I'm too old and cynical to believe it will work.

SiGraybeard said...

Thanks for the kind words about my ramblings. I think trying to figure out how we save our Republic is the big problem of our time. I don't have any grand kids, yet, but I never want to have to answer the question, "what the heck were you doing that was so important that you let liberty die??"

I think you're on to a really good idea. Remember Pareto: 80% of the savings come from 20% of the cuts (and vice versa). Cut the worst 20%, then re-examine and cut the next worst 20. In five cuts you're pretty much down to your lower left quadrant.

As you say, there will be push back. Everyone has a favorite program they want the Fed.gov to spend on, whether that's NASA, or Protect The Cuddly Polar Bears. Agencies that have outlived their usefulness, EPA comes to mind, need to be shuttered. Close BATFE first! What they need to understand is that if the government collapses economically they're all gone. Everything, every office, from the essential to the least useful.

My biggest fear is that it's too late to do anything and collapse is too imminent.

Anonymous said...

Speaking of which, if you haven't read Kelly Johnson's autobiography, you should.

Jim

Paul, Dammit! said...

The CFR's (At least the ones that are related to oil and maritime issues) are my bible, the end-all argument ender. Stupid or not, the history of the emplacement of each of these rules has a backstory. Some are compromissory in nature, others were paid in blood. The difficulty I see in this approach is that the fixes can't be solely either averaging mechanisms (useful in compromise situations) or reliant on a command-and-control style of management. Choosing where and when to alter your approach to improving the efficiency of a management scheme will mean using value judgements, and the CFR's are all-encompassing. The political gauntlet to be thrown down first, in my eyes, is who can and can't be a subject matter expert with the power to advise on changes, how to define and measure the benchmarks for where and when change is necessary, and how to define success, since the CFR's aren't purely based on economics, but also cover things like safety issues and delegation of responsibility for co-management of some issues.

I'm not 100% sure I see the methods above as being sufficient to guide in making changes- rather, this approach seems like an assessment tool to measure effects of changes... I absolutely could be talking out of my ass here, however.

GuardDuck said...

Every federal regulation down to the lowest municipal zoning book. Thats the kind of book burning I can get behind. Start from scratch.

"nuke it from orbit, it's the only way to be sure"

SiGraybeard said...

Paul's comment is important. I'm in the aviation business, and there's a cynical saying that every FAA regulation is written in blood. That is, things are not addressed until a major accident happens.

No one said it's going to be easy, but I'm sure there are efficiencies to be gained and improvements to be made by dumping regulations and laws.