Sunday, July 31, 2005

Prison Studies and Common Sense

My partner Mr. Turner, has done a good job of explaining PIECP as relates to the proposed prison. We have turned up some new goodies in that area, but some of our inquiries will not get answered until next week, we will update as we have more solid information.

I did get a copy of two economic impact studies commissioned by the Chamber of Commerce. Anyone who thinks that reading such material is dry, boring work just doesn't have the right attitude. Throw in a couple of adult beverages and my tour of these studies was a hoot. The one prepared for the State of Texas was a gas.

First, the cost of construction and equipment in the study lists at $17,780,000. Now the contract the Commissioners' Court is looking at specifies “cost of construction not to exceed [slightly under] $24 million”. I believe the last figure I saw in the Standard-Times was $31 million. Might be helpful if we had a clue what this “Pearl of the Concho” was going to cost.

In one of the tables, this study gives the total state sales tax revenue from workers as $162,408, and this is for a 2008 actual opening date. Sounds like they know, down to that last sack of Cheetos what these guards are going to buy three years from now.

On the construction phase the study has one table projecting, per worker, how much each worker will contribute to the state in tobacco tax, alcohol tax, and lottery tickets, $77, $75, and $187, respectively. Now it ain't hard to picture a herd of chain smoking, hard drinking, inept gamblers in hard hats, but then on page 23, the same table is repeated as an estimate for the guards at the prison once it's built. Sounds like a lot of backsliding gonna be going on amongst these exclusively born again Christian guards, but what the heck, prison guard is stressful work.

The job creation numbers are 153 at the prison, with 143 “indirect or induced” jobs. The 143 figure is slippery, these are the clerks (and presumably bartenders) who will be hired to accommodate the business brought in by the 153 real jobs. Now I might suspect that the nearby Town and Country store will have Louise stay over an extra half hour to help with the shift change rush instead of hiring a new hand, but then I do not have an MBA, so what do I know?

The line about “indirect and induced” reminded me of a phrase I caught in another study, this one a federal inquiry into drug use. Since the raw data was obtained by literally going door to door and saying “Hi, I'm from the government, do you use drugs?”, the pros would then “adjust for nonresponse by imputation”, seems they were unsure everyone would answer truthfully. In layman's terms this is called “making things up”. “Induced” is a red flag word in any study.

I really shouldn't be hard on this Impact DataSource company. Short of doing a very pricey direct survey, this guy did what is industry standard, he pulled a model from Dept. of Commerce, (in this case RIMS II) selected his multipliers and assumptions from a likely list, fed in the numbers provided by Chamber of Commerce and probably went for coffee while the machine hummed for a while and spit out the result. At the end of the study he freely gives a list of all this, nothing dishonest about it.

Unfortunately real people then take these results and make expensive decisions based on them, just as though they meant something real. The lesson here is, DO NOT take any “study” of anything as better than a slightly educated guess until you know the process by which the numbers were derived, and then keep several large grains of salt handy. When people get up and start throwing these numbers around, feel free to challenge them, they are next door to meaningless.

No comments:

Post a Comment