Wednesday, February 17, 2016

Do waivers work? Compared to what?

In the middle of the last decade, the State of Florida and two counties in California got a head start on the pack in developing and implementing the modern era of child welfare waivers. These waivers allow federal funds now reserved for foster care to be used on better alternatives, including family preservation strategies.

Key measures tracked in evaluating waivers include: entries into care, number of children in foster care on a given day, re-abuse of children left in their own home and re-abuse of children after they are returned from foster care.
So, how did they fare?
§  In Florida, entries into care declined significantly, and the independent evaluation of the waiver found that child safety improved.
§  In Alameda County, Calif., there was a big drop in entries into care, and both key safety measures improved, according to the California Child Welfare Indicators Project.
§  In Los Angeles County, foster care entries barely changed; the point-in-time number actually declined more slowly than before the waiver.  One safety measure stayed the same, one got worse.

The conclusion is obvious, right? Get rid of the waivers – they’re endangering children!

Only in the upside-down, inside-out world of American child welfare would anyone think that makes sense.
What also makes no sense is how we fund child welfare now. The giant, open-ended entitlement reserved for foster care dwarfs the meager amount of federal aid available to keep families together.
This does not mean, as some on the far right say, that governments make money on foster care. But it does mean that even when safe, proven alternatives cost less in total dollars, it can cost a state or a county less to place a child in foster care. And many private agencies, paid for each day they hold a child in foster care, do indeed make their money that way, creating a perverse incentive to prolong foster care.

Waivers are a very small effort to redress this imbalance, because anything bigger is routinely defeated by lobbying from agency trade associations such as the Child Welfare League of America.
So, how do we know waivers won’t encourage a mad rush to keep children in dangerous homes? It didn’t happen in Florida. And it didn’t happen in Alameda County.

It also didn’t happen in Illinois, which didn’t wait for a waiver and moved to change incentives for private agencies on its own. In 1997, there were 50,000 children trapped in foster care in that state, proportionately more than any other.
Then, instead of simply paying per diems to private agencies, Illinois started rewarding them for permanence – adoption and reunification – and penalizing those that didn’t deliver. Today there are about 17,000 children in foster care in Illinois, and independent court-appointed monitors found that child safety improved.

What about L.A.? The results tell us only that in one of the most dysfunctional, highly politicized systems in America – with, at the time of the waiver, possibly the worst governing body in all of child welfare – it’s going to take more than a few years to get a waiver right.

The Other Incentives
The other reason the Chicken Littles are wrong about waivers is that while financial incentives are an important incentive, they’re not the only incentive. All the other incentives encourage needless removal of children as well.
Personal incentives: When a worker sees a child living in poverty, the first instinct is often to “rescue” the child on the assumption that the child is bound to be “better off” in care.

The child in that impoverished home is a reality before the worker’s eyes. The dangers of foster care, physical and emotional, are an abstraction.  The problem is compounded by the class bias and racial bias that permeate child welfare. If keeping families together really is everyone’s first priority, as everyone claims, why is it that the only form of permanence most ever celebrate is adoption?

Political incentives: Many child welfare workers like to say they’re “damned if we do and damned if we don’t.”  But it’s not true.  Caseworkers and administrators don’t get fired, suspended, demoted or even slapped on the wrist for taking away too many children. All of those things have happened to workers and agency chiefs who “allowed” a child to die on their watch.

When it comes to taking away children, you’re only damned if you don’t. Indeed, one reason the waiver changed so little in Los Angeles was because of ongoing, inaccurate media coverage in the Los Angeles Times.  Something similar is happening now in Florida.  But while entries into care now are increasing again, they have not increased at nearly the rate they did under almost identical circumstances in 1999 – when there was no waiver.

Ideally, there would be no “external” incentives pushing child welfare decision-makers in any direction. But that’s impossible. So as long as so many incentives push workers to take the child and run, a change in financial incentives is needed to push back, just a little, the other way.
The “Compared To What?” Factor
The trashing of waivers also suffers from the same double standard that crops up whenever someone tries to change the foster care status quo. A few examples:

A whole lot of studies of Intensive Family Preservation Services programs that follow the Homebuilders model find them safe and effective, while a few studies of programs that don’t follow the model do not. Critics focused only on the studies of programs that don’t follow the model.

There are 26 studies of differential response, and 25 find that it does not compromise child safety. But one finds that that it does, so critics focus on that one and accuse everyone else of bias! (In fact, the only study I know of to look at scholarly bias in child welfare suggests the bias goes in a very different direction.)

And when one evaluation after another finds that waivers don’t compromise safety, critics focus on the one indicator from one evaluation that does.
That pernicious double standard allows things like the appalling rate of abuse in foster care itself to go unchecked.
We can’t measure how many Los Angeles children were saved from abuse in foster care itself by being sent home sooner. But given the number of studies that show abuse in one quarter to one third of foster homes (with a worse rate in institutions), it probably was a lot.

Even those studies are likely underestimates.  Read the scathing 200 page decision by a federal judge in Texas in a class-action lawsuit against that state’s system – a decision finding that abuse in foster care isn’t just common, it’s rampant – and tell me why we shouldn’t be demanding changes in the financial incentives that encourage needless placement.

The real problem with waivers is that they don’t go nearly far enough.  At a minimum, every state should be allowed to trade in its foster care entitlement funding for a flat amount of funding, adjusted for inflation, that could be used for prevention and family preservation as well as foster care.  Ideally, such a change would be mandatory.

Isn’t it time, finally, for the burden of proof to switch to those who want to keep on shoveling money into a foster care system that does so much harm to so many children?
This analysis originally appeared as part of a series in the Chronicle of Social Change series entitled: “Dollars and Priorities: The Financing of Child Welfare” where it can still be found - for the moment. Since the Chronicle, the Fox News of child welfare, has taken to suddenly moving much of my work behind a paywall, I'm reprinting it here, with some updates and changes to the links.