I could not vouch for the "six" but I doubt it is far off. It may even be an underestimate.
http://www.niemanwatchdog.org/index.cfm?fuseaction=ask_this.view&askthisid=0093 There was a dramatic 60 percent national decline in the number of people receiving federally-funded, welfare (TANF) between August 1996, when President Clinton signed the welfare reform bill into law, and December of 2003, according to the U.S. Department of Health and Human services. The number dropped from 12.2 million people to 4.8 million.
That's a little under eight million, I believe. However:
http://www.brookings.edu/views/speeches/waller/20041114.htm Furthermore, analysis of welfare caseloads suggests that program design or administration has already contributed to caseload decline. In the mid-1990s, over 80 percent of all families eligible for cash assistance got it, while in the most recent report to HHS, less than 50 percent were getting help.
So even in the mid-nineties, when an expanding economy made it possible for more people to find work, around 20% of those eligible for cash assistance were not getting it. If 12 million were getting welfare in '96, 20% would be over 2 million. So add those to the eight million or so noted in the first quote. That's ten million. Sounds good - ten million off welfare. The problem is that it has always been hard to track those who went off.
http://www.niemanwatchdog.org/index.cfm?fuseaction=ask_this.view&askthisid=0093 Government studies on welfare almost always fail to fully account for people who have left the public assistance rolls but did not obtain employment. For example, the most significant New York State study on this topic, “Leaving Welfare: Post-TANF Experiences of New York State Families, June 2002,” was able to obtain information from only 53 percent of sampled families.
The study assumes that the families interviewed had identical outcomes to the 47 percent of families who did not respond. But it is highly likely that the families that could be located for interviews had far better financial and employment conditions than those families who did not participate.
(by the way, this author worked in the Clinton administration and still "on balance" thinks he "did the right thing.")
If you extrapolate NY to the rest of the Country, approxamately half of the ten million who either left the rolls or didn't get the benefits they were eligible for in the first place = about five million. Not so far off six. Nor is that the total story.
http://www.niemanwatchdog.org/index.cfm?fuseaction=ask_this.view&askthisid=0093 According to more recent statistics from New York City, out of the people who left public assistance over the past year, less than 40 percent did so because they were placed in jobs. Out of that 40 percent, fully 24 percent were no longer employed after six months. Thus, almost 70 percent of people who left welfare in New York City had no post-welfare employment reported after six months. Because the City does not track former recipients beyond six months, we have no idea how many more former recipients may have lost employment later on.
Other data from around the country now indicate that, especially in this still-difficult economy, many people left welfare without having jobs, but even many of those who did obtain jobs early on likely lost them later, or kept jobs but didn’t earn enough to feed their families.
Now, you may accuse me of mixing statistics, or playing with numbers, but I assure you that is not my intent. It has always been difficult to find hard data on welfare, even before "reform." Different studies use different measures, etc. And I am no statistician, so simply do my best to wade through.
However, even BEFORE welfare "refrom" there was reasonably good data (at least to the best I could evaluate it) that most families cycled on and off welfare as the availability and feasability of work changed. The "rolls" were quite fluid - some families going on, some going off. People would get a job, then get laid off - or hours cut - or a child would get sick - or the car would break down - or the Day Care provider (often a relative or friend) would have to quit for some reason - or, or, or. Working is very hard when you're poor, just logistically.
Since you seen to think welfare "reform" worked so well, here's a little experiment for you. Borrow an infant who's 18 months or so overnight. Get up at 7:00 or so for your 10:00 AM shift at K-Mart. Feed and dress the baby, put together his/her supplies for the day, get out the stroller and walk the three blocks or so to the bus stop. (This is better is sub-zero winter weather, but I won't insist). Take the half hour ride to the Day Care. Get off the bus, walk the blocks to the Day Care home, get the baby settled, walk back to the bus stop, take another half hour or so ride to work. Work five hours, because they've cut your shift. Repeat the morning in reverse. When the baby is sick the next day, call in. You might lose your job, but the Day Care doesn't take sick kids. You still are not making enough even to pay your own rent, so you still have appointments at "the Welfare." Try to schedule them on your day off - difficult, because they make the appointments weeks ahead, and you don't know Monday's schedule till Saturday before.
Yes, there were problems with the welfare system before "reform." They could have been fixed without the draconian, punitive measures instituted. Most families wanted to work, tried to work, even before "reform." The initial success of "reform" was due largely to an expanding economy and would have happened anyway, I think. Other success was due to backward states which hadn't had support programs for poor working families instituting them (here in NY we did have them before "reform."). Again, those reforms could have been made without the draconian measures. We see the results of those measures now. In my area, approxamately half the Food Bank use is by low-income working families. Around the Country, our homeless, child-poverty, and child-mortality figures are appalling. The prisons are bursting at the seams. That's some real success, eh?