Sunday, January 30, 2005

The Definition of a Truly Libertarian Society

The definition of a truly libertarian society is surely one where escaping from prison is not a crime:

Prisoners Undercut Mexican Drug Crackdown (washingtonpost.com): "To the dismay of many law enforcement officials, escaping from prison is not a crime in Mexico. As long as the escapee does not commit another crime while escaping -- such as assaulting a guard -- there is no penalty. As one Supreme Court justice has explained, 'the person who tries to escape is seeking liberty, and that is deeply respected in the law.'"

Friday, January 28, 2005

Doubts about DHS New Personnel System

Civil Service System on Way Out at DHS (washingtonpost.com):

From Wednesday's Post
"The Bush administration unveiled a new personnel system for the Department of Homeland Security yesterday that will dramatically change the way workers are paid, promoted, deployed and disciplined -- and soon the White House will ask Congress to grant all federal agencies similar authority to rewrite civil service rules governing their employees.

The new system will replace the half-century-old General Schedule, with its familiar 15 pay grades and raises based on time in a job, and install a system that more directly bases pay on occupation and annual performance evaluations, officials said. The new system has taken two years to develop and will require at least four more to implement, they said."
I've reservations about such changes. The world little remembers the Civil Service reform efforts of Jimmy Carter, but maybe I'm not of this world. The Senior Executive Service was part of it (Not the world, Carter's reform). Never having reached that exalted level, I'm not sure, but I don't think it fulfills the promises made for it. Another part, now scrapped, was a reform of the merit pay system, less far-reaching than the DHS system but subject to some of the same problems:

  1. Do the personnel (sorry, human resources) people in HSD buy into it? That's not clear. "Not invented here" is a real problem everywhere.
  2. Conflicting priorities--if you're a line manager in HSD trying to focus on your mission, you may well resent diverting your attention to learning the ins and outs of some new fangled idea, particularly if your employees start worrying about the process and it looks to you as if the new system won't improve your ability to manage. People find reasons not to change.
  3. Reluctance to judge--I never saw a manager or employee who welcomed a discussion of strengths and weaknesses. Maybe reality TV shows like American Idol prove modern youths are different. Maybe Judge Chertoff can create a culture in DHS of frankness without recrimination. Maybe there's a tooth fairy. The Civil Service System is not bad, it just doesn't work. The words on paper are good, the actions in reality are something different. The challenge for DHS is to achieve some congruence between paper system (where "paper" includes "computers") and what happens every day.
  4. Lack of performance standards--one of the reasons we have government perform a function instead of free enterprise is the lack of clear standards.
Maybe the changes will work as claimed. More likely after 20 years or so we'll find there's some improvements and some problems. Or possibly, DHS and the reform effort will require more reforms, more reorganizations.

Wednesday, January 26, 2005

Big Software Projects Fail

Note the following is not unique to government. I firmly believe that organizations of people are more similar than different.

The FBI ran into problems with its big software project, leading to an op-ed in
The New York Times > Opinion > Op-Ed Contributor: Does Not Compute

"Research by the Standish Group, a software research and consulting firm, illustrates the troubled fates of most big software initiatives. In 1994, researchers found, only 16 percent were completed on time, on budget and fulfilling the original specifications. Nearly a third were canceled outright, and the remainder fell short of their objectives. More than half of the cost overruns amounted to at least 50 percent of the original budget. Of the projects that went off schedule, almost half took more than twice as long as originally planned. A follow-up survey in 2003, however, showed that corporate software projects were doing better; researchers found that the percentage of successful projects had risen to 34 percent."

I was involved in several such projects at USDA (i.e., large computer systems covering multiple offices and agencies). That led to my formulating Harshaw's rule one (you never do it right the first time). Other factors for failure:
  1. the culture gap between information technology (IT) (used to be EDP, then ADP) types and the operational types who really call the shots.
  2. McConnell's rule -- consider the goals of a project: cheapness, speed, quality. It's easy to achieve one, difficult to achieve two, and all but impossible to achieve all three. (I think I stole that from Steve McConnell's "Code Complete", published in 1993.) It fits my experience, although Dan Goldin, the former NASA head had the motto: Faster, Better, Cheaper.
  3. hubris. See the classic tragedies. See "A Bridge Too Far." See the Spruce Goose.


Tuesday, January 25, 2005

Harshaw's Rule One

Over my years in USDA I developed a set of rules for myself. The one I used most often is:

"You never do it right the first time".

My point here is that this is my first blog.

Richard Hatch and the IRS

Score one for the good guys. Richard Hatch, the first million dollar winner on "Survivor" is being hauled into court by the IRS for failure to pay taxes on his winnings. (http://www.thesmokinggun.com/archive/0118051hatch1.html)

I don't like what the economists call "free riders", so I'm happy. But the Washington Post's Richard Leiby (The Reliable Source column) played it for laughs, implying
Hatch was stupid not to realize that the IRS watches Survivior and checks up on winnings. He made an assumption (one of my pet peeves) which may be wrong. I hope it's wrong. I hope we aren't paying any bureaucrats to watch TV to catch winners of quiz and reality shows. Put aside the question of what would be a fair salary for such duty. (Full disclosure, I've never watched a reality show.) As a bureaucrat (retired) I think the fairest and most effective tax collecting system is just that, a system, ideally computerized.

If IRS doesn't have TV watchers, how was Hatch caught? Possibly an informer,
which may be hard to believe, as apparently he's the most lovable of men. But an informer might even pick up a percentage of the take (I'm no expert on tax law). I hope that wasn't the case--there may be instances where only an insider who turns his coat can catch a law violator, but they should be rare.

I hope what happened is the IRS systems worked. The network (or "Survivor Entertainment Group") included the payment in its report to IRS, along with all the other payments of salaries and bonuses that would qualify as income to the recipient. When Mr. Hatch's 1040 came in, I hope an IRS system matched it against the network report and found the discrepancy.

Of course the media has little interest in such issues, which can build public mistrust of our hardworking bureaucrats.

(Faceless bureaucrats aren't paranoid--they have real enemies.)