nor have emergency room visits for crack users diminished all that much. What did go away were the huge profits for selling crack. The price of cocaine had been falling for years, and it got only cheaper as crack grew more popular. Dealers began to underprice one another; profits vanished. The crack bubble burst as dramatically as the Nasdaq bub- ble would eventually burst. (Think of the first generation of crack     dealers as the Microsoft millionaires; think of the second generation as Pets.com.) As veteran crack dealers were killed or sent to prison, younger dealers decided that the smaller profits didnt justify the risk. The tournament had lost its allure. It was no longer worth killing someone to steal their crack turf, and certainly not worth being killed. So the violence abated. From 1991 to 2001, the homicide rate among young black men-who were disproportionately represented among crack dealers-fell 48 percent, compared to 30 percent for older black men and older white men. (Another minor contributor to the falling homicide rate is the fact that some crack dealers took to shooting their enemies in the buttocks rather than murdering them; this method of violent insult was considered more degrading-and was obviously less severely punished-than murder.) All told, the crash of the crack market accounted for roughly 15 percent of the crime drop of the 1990s-a substantial factor, to be sure, though it should be noted that crack was responsible for far more than 15 per- cent of the crime increase of the 1980s. In other words, the net effect of crack is still being felt in the form of violent crime, to say nothing of the miseries the drug itself continues to cause.       The final pair of crime-drop explanations concern two demographic trends. The first one received many media citations: aging of the popu- lation. Until crime fell so drastically, no one talked about this theory at all.