A Conversation with David Ellwood

After the Safety Net: A Welfare Reformer Reflects on What Washington Wrought

David T. Ellwood is the Lucius N. Littauer Professor of Political Economy and former
academic dean at Harvard's John F. Kennedy School of Government. He previously served as
assistant secretary for Planning and Evaluation, Department of Health and Human Services.
After serving as one of the two chief architects of President Clinton's welfare reform proposal,
Ellwood resigned in protest over Clinton's support of the Republican bill -- a bill that did not
provide jobs or other protections for welfare recipients who reached welfare time limits without
finding a job.

Q: You were one of the principle architects of a welfare reform policy that
pledged, "Two years, and you work," with some sort of job guarantee if no
jobs were available. As you wrote in this magazine in 1996 ["Welfare Reform
as I Knew It," TAP, May-June 1996
], Congress changed that to "Two years, and
you're off [welfare]" with no job protections. What harm has come of that change?

A: So far it hasn't made a whole lot of difference, largely because the
economy has been outstanding. We did some things right prior to and along
with the passage of welfare reform, one of which was to dramatically expand
tax credits to working families. We really improved the rewards to working
people who can find a job. We made work pay and we did welfare reform
in a great economic environment. So in some sense, the issue of "two years,
and you work" versus "two years, and you're off" is mitigated when there are
plenty of jobs in the private economy. The real test is twofold: What
happens in the next recession, and what about the folks who can't find jobs
even in a good economy?

Q: The public perception of welfare reform so far seems to be one of
cautious optimism. Will that change once more recipients reach their
lifetime limit and are forced off the rolls?

A: I'm not sure it will. So long as the economy remains good and people are
doing reasonably well, I suspect the cautious optimism will continue.
Partly this is because those who are the most desperately poor are often
not very visible. I think there will be some perception that there are some
folks being harmed, but I'm not sure the public perception is likely to
change, barring some major economic changes.

Q: It's been argued that we're in the midst of the "easy part" of welfare
reform, that case loads are dropping with little pain or public outcry. The
other side of that argument holds that once recipients start reaching
limits en masse, we'll experience the "hard part" of reform and its
accompanying problems. Do you put credence in this outlook?

A: I think that's right. I've always had a rough rule of thumb: One-third
of folks on welfare, with appropriate support, would leave welfare fairly
readily; a second third, with effort and energy, could be gotten into work;
and a final third, for whom the problems are extremely serious -- people who have limited education, little work experience, seriously ill children, or
have been subject to abuse and neglect growing up -- [would have a very hard time holding steady jobs]. Where we are right now is somewhere in the middle of that second group. Case loads are down about 50 percent, but already we're hearing from states that the people who are left will be a much more difficult problem.

Q: The liberal postmortem seems to be that Clinton's original welfare
reform proposal, which you helped craft, was quite good. Where did the
administration go wrong?

A: The Republican takeover of Congress really did change everything. The
president wasn't willing to fight for the original conception of "two years,
and you work" and indeed was comfortable with a more pure devolution and
strong encouragements to move folks off welfare very rapidly without
necessarily getting them jobs. The loss of the House and Senate [for Democrats] made it a much more difficult situation. The president, being a [former] governor, adopted a
strategy that pretty much lets states do what they want. I'm very
disappointed because I think what was lost in this process were some
protections for people for whom work is going to be very difficult.

Q: Would you do anything differently if you could go back?

A: I would try and move more quickly. Waiting toward the end of the second
year was really quite unfortunate. I don't think we in the administration
did nearly enough work with the Democratic Congress to make them
comfortable. There were some who had misgivings about the original
proposal. We made some mistakes with how this was portrayed in the media.

Q: Do low-income working families get a better deal today than they did
before welfare reform?

A: Absolutely, although it's not principally due to welfare reform. Those
of us who worked on welfare reform had a three-part agenda. The first part
was to make work pay, and a series of things have really improved the
situation for low-income working families, including dramatic expansions in
earned income tax credits, much more generous medical care supports for the
children of working parents, and more generous child care allowances. The
child-support enforcement system was also dramatically improved. What was
lost was protection for people who are willing to work but unable to find
reliable work. For us it was "two years, and you work" not "two years, and
you're off."

Q: One facet of your reform strategy was to change the welfare culture from
one of check-writing to one where work was the expectation and the goal
from day one. To some degree, that has occurred. What still needs to be
done to fix welfare reform?

A: The culture of welfare offices really has changed. I don't think there
are many offices anywhere that are primarily about eligibility and check-
writing. But I'm not sure that most welfare offices are about finding
people jobs, either. They're about saying, Listen, you've got to get going,
you can't get on welfare. The biggest issue with welfare reform is what to
do about the people who can't find work or for whom full-time work is

Q: Has the reduction in case loads been commensurate with the reduction in need?

A: It clearly hasn't. The case loads have gone down very dramatically and
poverty has gone down only somewhat. There is some evidence that people are
worse off at the very bottom. But keep in mind that we're spending more now
on social policy than we were at the start of welfare reform because of
this dramatic expansion of the Earned Income Tax Credit (EITC), medical support,
and so forth.

Q: Critics have long suspected that the goal on the right is to end federal
spending on welfare. Are the sharp case load reductions the first evidence
of this happening?

A: I really do think we're gradually shutting down the welfare system. The
basic support for people who are not in the disability programs and who are
not working is shrinking, and likely will continue to shrink into the
foreseeable future.

Q: You were heavily involved in Clinton's expansion of the Earned Income
Tax Credit. Today, far more money is spent on the EITC than is spent on
welfare. Do you think Americans realize how much money is being spent on
the EITC? And if so, do they support this spending on the poor?

A: That's correct; we now spend more on the income tax credit than we ever
spent on Aid to Families with Dependent Children (AFDC) or Temporary Assistance for Needy Families (TANF). But I don't think most Americans know how much is
being spent on AFDC, although I think that's true of a hundred other tax
credits as well. This is one of the Clinton administration's major
accomplishments that hasn't gotten the attention it deserves. We have
dramatically increased supports for the working family, and often in a
bipartisan way.

Q: You've noted how drastically an election -- even an election year -- can
influence welfare policy. How might a Bush or a Gore presidency affect
welfare reform?

A: That's a very interesting question. Based on what people have said so
far, I'm not sure it would make a huge difference on welfare policy. I
think Gore would do more to support working folks, but I'm not sure the
differences are huge on welfare per se. Both administrations are going to
have to address the problem of these hard-to-serve cases, once we're down
to that, though I suspect Gore would be more concerned about the issue. And
if there's a recession, both are going to have to deal with the fact that
the current system is not very well designed to cope with recessions. Under
either president, old-style welfare is unlikely to return. But I worry that
if we had a Republican Congress and a Republican president we might see
much more radical changes: perhaps time-limiting food stamps or the
elimination of other major parts of the safety net that could be very, very

Q: What has happened to those pushed off, and what will happen as more are?

A: We know that about 60 percent are getting jobs. No one is quite sure
what has happened to the rest. Very few people relied solely on welfare, so
there's the food stamp program, which remains in pretty good shape, housing
programs, and state and local programs that help some people. But I think
there is a group of people who are clearly becoming worse off whom I'm
troubled about. That remains an issue.