by Tom Corbett
When Bill Clinton tapped Donna Shalala, then chancellor of the Madison campus, to head Health and Human Services, she asked a few IRP affiliates over for a briefing before heading to Washington. As we discussed several salient issues, she was surprisingly open about the challenges she faced and her lack of substantive knowledge regarding many of the major issues on the table at that time. We were all confident she would do well, however, perhaps more confident than she was.
Though diminutive in size, Donna Shalala had a powerhouse personality. After all, when she came to Madison, the Athletic Department was in the crapper. In a few short years, Wisconsin was turning out powerhouse football and basketball teams. She knew that was the way to open the wallets of alumni and it worked. Her last words the day of our briefing were that, once in office, she would beef up ASPE which had diminished in size and importance over the years.
ASPE had been created in the 1960s as a long-range planning arm for the HHS Secretaries office. It was also supposed to provide a centralized research and analysis arm to coordinate thinking across the department’s broad programmatic responsibilities. In addition, it was supposed to pick up many functions that had been in the old Office of Economic Opportunity (OEO), which President Johnson created as a command center for his War-On-Poverty (WOP). The WOP was waged from the White House from the time that Johnson made poverty a public issue up to the point where President Nixon turned to other priorities. It was also hoped that ASPE might crosswalk among the diverse program areas within the mega-departments purview, which at the time included education issues, to achieve a modicum of systems integration.
By the late 1970s, ASPE was a power place to be. I happened to be present when Henry Aaron, a Brookings Institute scholar of iconic stature, told David Ellwood, the Harvard scholar heading ASPE early in the early Clinton era, about the relative strengths of ASPE then and now. When he (Henry) had headed the place during President Carter’s push for welfare reform, his staff was top-notch and probably could compete with a top university economics department. Henry then told David that the current staff was not up to the job. Gary Burtless, another Brookings scholar of repute who had been at ASPE back in the 70s, nudged Henry and said he was insulting several people in the room. Henry didn’t miss a beat and carried on about the dire fate that awaited poor David.
It was true. ASPE was smaller and the staff less accomplished overall. But I found them to be generally competent, disciplined, and hardworking. Besides, David Ellwood and Mary Jo Bane (another Harvard professor who was heading ACF or the Administration for Children and Families) were bringing on board quite a few academics and analytic types. Unlike me, however, most of these other academics had real number-crunching skills.
In any case, we plunged into the work of “ending welfare as we knew it.” Cross-agency work groups were setup to be responsible for various substantive issues. Initially, I was asked to head the “make work pay” effort. Basically, this looked at how various welfare-type programs could be massaged to incentivize work properly. This was something that welfare, by definition, did not do. But I also got involved in the program integration work group and did a little work with the child support work group. I was from Wisconsin, after all, the birthplace of the recent child support revolution. I also plugged to have attention paid to the population I called FUBARS (fouled-up beyond all recognition), a term I am embarrassed to say caught on among my new colleagues. My short attention span always pushed me to jump from one issue to another.
The FUBAR issue was critical in my mind, despite my unfortunate choice of labels. The framework for the Clinton initiative presumably emerged from David Ellwood’s well-received book titled Poor Support. In the book, he argued for a temporary form of cash assistance, shocking only because he was considered a classic liberal. Basically, a client could get up to two years of cash assistance followed by a job. If they couldn’t get a real job then a public job would be made available, the details of which were formidable. If they didn’t respond to that, things got just a little vague.
Everyone knew that some unknown portion of the welfare population would be a real challenge, having very low skill levels and significant behavioral and mental health barriers. I thought we should jump on that issue, try to get an estimate of the size and composition of the population. Then we could think hard about what to do with them. Simply casting these families off assistance seemed ill-advised and a bit harsh while simply exempting them from any time limit lacked imagination. Interpolating from several different data sources, we came up with a 20 percent estimate which, drum roll please, became the exemption rate that later was built into the bill. I, however, wanted us to think about doing something more proactive with this group than continuing to issue an assistance check. For some reason, that never rose to the top as a priority however. Most feared that getting even employable clients into the labor market would sop up existing resources, leaving nothing for the unemployables.
Over time, I wondered who had come up with the “end of summer” target date to begin with. There were so many moving targets to this effort that simply communicating among stakeholders and coordinating the various tasks became herculean challenges. Ann McCormick had an office next to me. I thought Ann a treasure with her quick Irish smile and a similarly quick mind. Among other things, she was put in charge of keeping the paper work flowing to and from the extended set of people within ASPE and across all the related agencies and actors that needed to be in the loop. This seemed like half of Washington at some point.
Anyways, it struck me as odd that Ann M. had been assigned this responsibility. She had the second messiest office that I ever had seen in my life. When I peeked in her office, all I could see were piles of paper. I would call out, “Ann, are you there.” Then I would wait for a voice to emerge from the other side of a wall of paper. A terrible thought hit me, maybe everything went in and nothing came out. My fear proved groundless, she did have a method to her madness. In case you are wondering, I am not first but third on the list of offices to be condemned by the Board of Health.
In first place was Don Oellerich. Don and I had overlapped in the Social Welfare Doctoral program at Wisconsin though he managed to finish about a decade or so before I did. The betting had been that I would yet be an ABD type (all-but-dissertation) long after I was sitting around drooling in my nursing home rocking chair. After a few years in academia, Don found his niche in government. He was a bright star at ASPE and would become the chief economist at Health and Human Services even though he had no (or very little, at least) formal economics training. As the years rolled on, Don has been credited by everyone who cares as the one person most responsible for saving the university-based poverty research centers.
Don was one of those thin, intense, driven guys. In that sense he was the opposite of me. On the other hand, he was like me in one important sense. He had a very messy office, surely messier than mine, even messier than Ann M’s. Since I interpret messiness as a sign of genius, he is up there with Albert Einstein, Sir Isaac Newton, and Aristotle. Still, he knew where everything was. “Don, you got the analysis CBPP did on marginal tax rates?” He would pause, call in a forklift to dig a path to the far reaches of his office, then excavate down three or four feet of paper and reports before pulling out what I needed. My hero!
Though I found the contest for messiest office a welcome diversion, I started worrying about the prospects for reform. It was soon apparent that “ending welfare as we know it” may have been a signature campaign line but it was an in-house powder keg. Fault lines could be seen but generally were kept under wraps, at least for a while. Though generalizations are inevitably faulty, overarching normative positions were easy to detect. The White House representatives involved in the everyday planning, particularly Bruce Reed, took a harder line. There had to be real time limits, for example. The president had promised to “end welfare as we knew it” and that is what we were going to do.
The other main actors generally took softer lin
es. People from Labor, headed by Robert Reich during this period, tended to be much more liberal in their positions. We could never push people off a cliff…something had to be there to catch them like a guaranteed job. Failing that, they would keep getting a check. What kind of position that would be, and what would happen if they failed at the job were continuing sticking points. Ending the welfare discussions was proving as difficult as ending welfare itself.
Not surprisingly, the key actors within HHS came from different camps with David Ellwood, our Harvard-based academic, trying to straddle the fence. His book called for time limits, but he now faced the reality of the intellectual position he had staked out. It is a lot easier to write about an issue than make decisions that affect the lives of real people. I blithely wrote on many sensitive policy issues absent concern since I assumed that, in the end, no one would listen to me. I suspect he had hidden behind the same assumption. Now, however, his emotional response to the reality of a welfare time limit seemingly wavered somewhere between ambivalence and uncertainty.
Mary Jo Bane, another Harvard transplant, was a little harder to read. As someone who ran a program in the real world at one point, I felt she was more cognizant of the challenges associated with moving away from a business-as-usual posture. Wendell Primus was the clear liberal of the HHS leadership group. It struck me that he was desperately trying to hold back a tidal wave of change he saw overturning the safety net about which he retained considerable passion. Still, it was hard for him to be totally open about his preferences since they might well be construed as being disloyal to the top man, Bill Clinton.
Few major policy issues are debated without a range of opinions. Perhaps the range in this instance was a bit broader than most. Strangely enough, I never heard Donna Shalala’s name mentioned at all. Based on her strong presence while chancellor in Madison, I found this puzzling. In my opinion, she was never comfortable with this part of Bill’s agenda and preferred to focus on health care. There was another group, composed of cabinet secretaries that occasionally would meet on welfare reform. Their ruminations, however, seemed ethereal to me and had no substantive influence on the planning process as far as I could see.
The real work was done by grunts like me. There were many meetings that started at 8:00 a.m., very early for D.C. The cast of characters would differ from meeting to meeting but David Ellwood, Mary Jo Bane, Wendell Primus, Bruce Reed (representing the White House), Belle Sawhill (a Washington fixture), and various reps from Labor, Treasury, Education, Agriculture, Housing, etc., usually would be in the room. After a hundred or so of these sessions that I managed to attend, I came to realize that we were not getting anywhere.
I started referring to them as the Groundhog Day meetings. This appellation came from the Bill Murray movie of the same name in which Bill lives the same day over and over. That scenario was a bit of a nightmare that I now felt was taking over my own life. In retrospect, what happened is not difficult to infer. The top players with real juice (David, Mary Jo, Wendell, and Bruce) were aware of the ideological divides that separated the main players. I doubt they wanted to display these fissures openly in front of staff. So, they kept the planning meetings focused on technical details that were hard but less consequential. But without the overall vision, progress was slow and not going in any positive direction as far as I could tell.
As a result, we spent tons of time on issues such as how would we run the clock that calculated the two years that clients were permitted to receive welfare? Under what conditions would it stop, or start, or even start over? I remember one day when we seemed to be arriving at a conclusion on this one issue. Since I had done this welfare migration study, I was well informed of the interstate movement patterns of welfare clients. Reluctantly, I spoke up. “Aaah, now what happens if a client moves across state lines? What happens to the clock? Let us say she sees her clock running out and decides it is time to move from Chicago to be with her sister in Milwaukee, a move that happens more often than you might guess. Does her clock start over? How would the case worker know about her clock status if she came from another state, from Illinois for example? Would they have to call other states to check some states or all?”
I stopped talking as all eyes bore in on me. I had this sudden vision that they were considering placing a note to my chest saying “please feed and give ride to Madison, Wisconsin” before chucking me out the window to the street below. “Okay,” David said slowly, “we need a national data base. It can’t be that hard, I probably could program one myself.” I looked out the window to see if any pigs were flying by.
What was not being dealt with were the fault lines simmering under the surface. They reflected political perspectives on one level and foundational norms on another. If you say you will “end” welfare, you better do so, or Republicans would bitterly attack the president for reneging on such an overt promise. On the normative level, some of the more liberal among our happy family had been defenders of the poor most of their lives. They found biting this bullet of ending cash assistance as an entitlement almost unbearable. Wendell Primus was a perfect example.
Wendell was one of the nicest individuals I have ever come across. He was also one of the most hardworking. A Midwesterner, he brought his heartland work ethic to his Congressional staff jobs on the Democratic side of the aisle. His reputation was that he would out negotiate everyone around the table, enduring long after his opponents dropped to the floor with fatigue. He brought the same passion to this ASPE position as one of Ellwood’s deputies. I would get there before him in the mornings, but he would be there in the office until way into the night, at least his emails would continue to arrive timed at midnight and beyond.
There was a frisson of shock throughout ASPE one day as a rumor spread that he would be taking a long weekend with his family. Betting was 10-1 that this would never happen and still no one was taking the bet. His closest staff had to literally push him out the door while he continued to shout instructions to everyone in sight as his wife and family waited downstairs. Wendell had spent his whole life defending the safety net for the poor. Ending it was almost impossible for him. I am sure he worried what might happen if he abandoned his post, and the barbarians from the right stormed the citadel.
There were, of course, many other impediments to progress. Take, for example, the simple issue of paying for reform. We were in a “pay as you go” environment where new fiscal outlays had to be offset by defined budget cuts elsewhere. Guess what, welfare reform cost money. If you were serious about getting people into jobs, there were training costs and child care costs and the costs of those guaranteed public jobs if enough private jobs were not available. The scramble for offsets was furious and futile. Well, someone would suggest we could raise sin taxes on gambling winnings. That idea would last a day until the governors of New Jersey and Nevada and several Native American tribes circled the Humphrey Building where HHS was located. Okay, who has the next idea? That challenge never went away.
There were the usual turf issues. The safety net is spread over several executive agencies and Congressional oversight committees and subcommittees. At a minimum, you had Health and Human Services, Labor, Agriculture (Food Stamps at the time), Housing and Urban Development, Education, Treasury (the EITC), and possibly Commerce, and a few others. I was even sent over to substitute for Ellwood at meetings that included high-level officials from the Defense Department. I can’t quite recall the relevance to what we were doing though the meetings were always held in the evenings in the Old Executive Office Building (technically the White House). Even in those days, pre-9/11, it was hard getting into that place…like you needed a secret password or such. Late one night, I left the meeting and got lost trying to find the one egress still open. I wandered up to a door labeled “Office of the First Lady.” Certainly, I was moments away from being pinned to the ground and being subject to a body cavity search when I finally was able to make my escape.
The point I was getting to, before my ine
vitable digression, was that too many fingers were playing in the kitchen. The safety net had evolved over decades with little thought to how the parts could possibly work together. By the 1990s, doing comprehensive reform had become a nightmare. You had to get all these agencies, each headed by turf-conscious leaders, to collaborate on a new direction and a new set of rules. Hell, it was almost impossible to get the agencies to agree on definitions for basic things like what defines an assistance unit or how to calculate income and assets. The bottom line was that yes, it was theoretically possible but not bloody likely. It was more likely that the next time I flew back to Madison for a weekend visit home, I would be hopping on a magic carpet for the flight.
Oddly enough, there also were failures of intelligence. I don’t mean that the planners were stupid, they were very smart indeed. Besides, they had access to the best and brightest from academia, think tanks, and the top evaluation firms. You could call, and people would come running, many of them at least. A lot of people wanted to be a player in this game. The problem was subtler than that. In some respects, it was the research itself that had led us astray. This is a statement which, coming from an erstwhile member of the academy, amounts to heresy. So, let me explain in some detail.
We had sound research methods to test small to medium-size research questions. The classic experimental design was perfect when you had interventions that were well defined and where subjects (individuals, families, or discrete bureaucratic units) could be randomly allocated to experimental and control groups with relative ease. If done well, this could control for alternate explanatory factors. But when the interventions were large scale and involved, for example, changing the overall culture of a program or interrelated set of programs, researchers could seldom, if ever, control enough of the real-world environment to maintain the rigor that a classical experiment demanded. You could hardly select experimental counties at random and walk in to say, “Okay, you folks are going to upend your whole world in the next six months, are you with us?” Economists hoped that by using large data sets and fancy econometric methods, they could tease out causality and control for those darn alternative explanations of any effects measured.