But there are other reasons why people are put in jail and kept there, over and above straightforward economic calculations. When my wife worked as a prison librarian in Connecticut in the 1970s, she had two inmate helpers named Stosh and Phil. Well, shortly after World War II, Stosh had cut his wife up into little pieces, and Phil had murdered a highway patrolman. And all those years later they were still in jail. So there are punitive, as well as economic, reasons for long prison sentences.

Crime data (left) suggests that California's expenditures are paying off. In 1997, for the first year in history since the FBI began keeping these statistics, the crime rate in California was lower than the national average. (In 1980, by way of contrast, it was about 30 percent higher than the national average.) Crime in California has gone down by about 35 percent over the last five years, as we've locked up very large numbers of people, compared to a decline of about 15 percent for the rest of the country. So we are locking up a lot of guys (left); crime is going down. Whether A causes B, who knows?

Moving on, I also charted spending on public assistance (bottom left). We historically have spent a lot more on welfare than the other states, but in the early '90s they caught up with us. They responded to the recession by increasing welfare expenditures, as one might expect. We didn't do that, as you may recall—we cut back on welfare spending instead. That is, California decreased individual payments. But the caseload grew, so the overall welfare budget held steady. In other words, the total pie stayed the same, but there were more people, so that each person got a smaller slice. In the last couple of years (not shown on that graph), expenditures nationwide have declined as welfare reform has kicked in and welfare rolls have shrunk dramatically.

And finally, I looked at spending on county hospitals and public-health programs. As you can see at the top of the next page, we kept pace with the national average pretty closely until the '90s, when we started spending about 2 percent more on health and hospitals. I don't know exactly what happened, but I think it's a substitution effect. People on welfare automatically get Medi-Cal, which is a form of health insurance. But when they leave welfare, they generally get low-paying jobs that don't include health benefits. Then, when they show up at the hospital uninsured, the county eats the cost of their care instead of charging it to the Medi-Cal budget. If you add the numbers together, you'll actually see quite a continuity in overall welfare, health, and hospital spending. It's just that we've shifted money away from the welfare account, which includes Medi-Cal, and toward the county hospital and public-health budgets.