<< Blog Scan | Main | The Validity of Internet Addiction >>

Spinning Corrections Spending Stats

CorrSpendPct.jpgThis post at SL&P points us to this story at stateline.org with ominous-sounding statistics about corrections spending. A closer look at the numbers tells a different story.

"Prison spending increased 127 percent from 1987 to 2007...." From the Pew Center's "One in 100" Report, page 12, it appears that this figure is adjusted for inflation but not for population growth. So the more relevant figure of inflation-adjusted dollars per capita is an 83% increase. That is still quite a substantial increase, but not as dramatic as more-than-double.

But is it true that the corrections budgets are the dominant factor in state financial woes, as many in the soft-on-crime camp would have us believe? No.
"Nationally, corrections trails only health care, education and transportation in consuming state dollars." Oh, my! Corrections trails only three categories of spending! Um, how many categories are tracked? The National Association of State Budget Officers, cited as a source in the story, tracks six categories, and education is split into two of them. The categories are (1) elementary and secondary education, (2) higher education, (3) public assistance, (4) medicaid, (5) corrections, and (6) transportation, plus a catch-all of "all other." Of these categories, public assistance is the smallest, and corrections is the second smallest.

So, of the six discrete categories tracked, corrections is actually fifth, trailing "only" four. See Table 3, page 8 of the 2007 Expenditure Report. Also, the "all other" category is 33.6% of total spending, nearly 10 times the 3.4% corrections share, so one or more of the "substantial" items in this amalgamation (see page 72), may well be actually larger than corrections.

at least five states -- Connecticut, Delaware, Michigan, Oregon and Vermont -- now spend as much or more on corrections as they do on higher education...."  Really? That's not what the NASBO report says (Table 5, page 10). Connecticut spends 11.9% of its total on higher education and only 2.8% on corrections.  The same numbers are 4.5% v. 3.2% for Delaware and 11.5% v. 3.9% for Oregon. It's true for the Green Mountain State because those famously stingy Vermonters spend only 1.9% on higher education, less than a fifth of the national average. Whatever the reason for that, it is a big stretch to blame it on the mere 2.4% they spend on corrections. Only in Michigan is corrections spending both higher than higher education and even a twentieth of the budget, according to the NASBO report.

Now for the most important number of all. How much is this rapidly growing monstrousity of corrections spending growing as a share of total state spending? Listening to the doomsayers, you would think it is gobbling up a larger and larger slice of the pie with every fiscal year, threatening to become the dominant item in the state budget. Right?

Wrong. The corrections share of the total budget is not growing at all. The graph at the top of this post simply the data from NASBO's Table 3 in graphic form. (Click on the thumbnail to get a larger version.) Corrections spending was 3.6% of the total in FY 1995, and it was 3.6% on average from then to the present.

Should we be taking a hard look at corrections spending? Of course. We should be taking a hard look at all government spending. Are there some sentencing laws that are too harsh and need to be reduced? Definitely. Vice-President Biden's 100/1 crack/powder ratio should be the first out the door.

But should we be panicked into rash action? No. The stateline.org story says, "The statistics are alarming state lawmakers in all regions of the country and, increasingly, on both sides of the political aisle." Well, they shouldn't. Concern, yes. Alarm, no. Panic, definitely not.

As we have documented before on this blog, the tough sentencing of the 1980s to the present has served America well. Tens of thousands of people have not been robbed, not been raped, and not been murdered because we saw the error of our hand-wringing 60s ways and cracked down. State laws against the basic crimes of murder, rape, robbery, burglary, and theft are not too tough, in general. The last thing we need to do right now is release the miscreants who committed those crimes early, so they can go out and repeat them.

Update: John Gramlich, author of the stateline.org story, responds by email, pointing me to p. 16 of the Pew report as the source of the claim that 5 states spent more on corrections than higher education in 2007. That is indeed what the Pew report says, but Pew's source is "reanalysis" of the NASBO report. No reanalysis is needed. The NASBO report states the facts flat-out, and they contradict the Pew Center's claim. The Pew report also has "reanalysis" figures for corrections as a percentage of state general fund spending on page 32, but even those "reanalysis" figures do not indicate that corrections is a rapidly growing share.

For those who would rather look at corrections spending as a percentage of general funds instead of total state expenditures, without "reanalysis", Table 3 of the NASBO report has those data as well. The percentage is nearly ruler-flat at 6.9% plus-or-minus 0.3% for the FY 1995-2008 period.

The Pew Center 1-in-100 report is very clearly an agenda-driven product. When the straight figures from NASBO did not support the agenda, they had to "reanalyze" the numbers in some unexplained way to make them sound more dire than they really are.


I wonder how much of that increase was for capital expansion? We built a lot of prisons over the past 20 years, so it's likely that this 20 year period includes some extraordinary expenditures for which we are still reaping benefit.

I wonder too how much the general increase in medical costs (independent of the aging of the prison population and the increase in the total number of prisoners) has affected expenditures.

It is amazing the lengths to which some will go to argue against common sense. Locking up criminals reduces crime and is a necessary function of the state.

Kent, what's your view on fixing the 100/1 disparity? To me, and I admit I am not fully informed on this point, crack and meth seem like pretty equivalent drugs (with meth being slightly worse given the dangerousness of the mixing process). Should crack be treated like meth for sentencing purposes, or should it be treated like cocaine, or somewhere in the middle?

Some differential between cocaine base and cocaine may be in order, but given how easy the conversion is, imposing a greater sentence for 5 grams of crack than for 499 grams of powder is way beyond proportionate to the offense. Similarly, "cutting" a drug by mixing it with inactive ingredients may warrant some enhancement, but treating a mixture the same as the pure drug of the same total weight is absurd.

Leave a comment