Sorting Through the DTA Audit

The Deval Patrick administration and Suzanne Bump’s state auditor office are in a big public pissing match over something that, as far as I can tell, everybody is essentially in complete agreement about. I spent my Friday trying to figure out why, and the result is this very long and, I suspect, poorly constructed post about it. I hope you might find it edifying, and if you hang in to the end you’ll get my speculation about why the Patrick administration appears to be bollixing this nothingburger into a cornucopia of nastiness.

As you may know, the state Auditor put out a report Tuesday saying that, from roughly mid-2010 through Fall 2012, the Department of Transitional Assistance (DTA) could, and should, have had better controls over its databases of public benefits recipients.

This is a thoroughly uncontroversial finding. The same has been declared by the state Inspector General, the federal government, and ultimately the Patrick Administration itself when it heaved DTA commissioner Daniel Curley out the door in February of this year. It is precisely why John Polanowicz,  the new Secretary of the Executive Office of Health and Human Services (EOHHS), moved his chief of staff Stacey Monahan into that commissioner position to tighten things up; and it is why she implemented a “100-day action plan” — heartily endorsed by Bump — that appears to be ahead of schedule in getting proper controls in place. (It is also why Monahan is currently advertising for a whopping 130 jobs, to staff up from the underfunding and overburdening of the recession years.)

So, I would have thought that the administration’s reaction to the audit would be to say, in effect, We thank the auditor for adding further detail to the assessment of the DTA’s prior problems; we are glad that she has endorsed our ongoing and ahead-of-schedule reforms; and we look forward to proving to her at the next DTA audit how much improvement we have made.

Instead, Polanowicz and the governor have publicly questioned the integrity of the report, and demanded that the auditor hand over her data; and, she has refused to do so.

It’s worth noting that this kind of dispute is not uncommon with public audits. (Or, in fact, private organizational audits, which I used to write about way back in my journalistic youth, although there’s a whole different level when it’s playing out in the press.)

Bump, Monahan, their respective flacks, and others gave me way more time than an obnoxious journalist has any right to expect on a summer Friday afternoon, to help me try to referee this statehouse clash. (Which all sides insist, unconvincingly, is not a clash at all.)

Clearly the administration side feels that Bump fed a false media narrative of massive ongoing waste and abuse justifying various legislative reforms. I think to some extent they have a point; the likely over-reaction of press and pols was easily predictable, and I think Bump’s team could have tried to mitigate that by taking more care in the presentation of the findings, which in some cases I think were not as qualified as they could have been. I would have liked to see some benchmarking of how other, similar agencies fare, and perhaps a little less emphasis on theoretical monetary losses that, while common and worthwhile in an audit of this type, are not partciularly relevant or justifiable. I don’t agree with those who have accused Bump of grandstanding, but I do think that the report and press release, and some of her public statements, overstated some points or left out useful caveats.

But, that said, the reality is that no matter how Bump wrote it or talked about it, many in the rabid right and the hyperventilating media were going to squawk about dead people receiving benefits, and scream for overreacting measures. They always do that. It’s not really Bump’s job to pre-overreact to predictable overreaction. It’s certainly not her job to help roll the audit out in a way that tells the administration’s preferred narrative of recent progress. Her office underwent, for two years, an important audit of the controls and data integrity at a major handler of public moneys, and reported the (again, essentially unsurprising and unoriginal) results.

The administration side is poking back at a number of specific findings, most especially the notorious numbers: 178 and 1164.

The auditors claims to have found 178 cases of recipient dependents — children receiving benefits via their parent or guardian — who came up as matches for being dead or using a deceased person’s Social Security number.  This is one of the cases where, particularly in the press release, the Auditor in my opinion should have been more careful with the language; it strongly suggested that these were all dead people receiving benefits, when in fact they were instances that seemed to clearly require investigation to find out whether that was the case. From the public’s perspective, these are very different things, although from an auditor’s perspective, the point is the same: your controls as of last year were clearly failing, and needed improvement. (Improvements that, as I mentioned, have been and are being implemented.)

During the audit/response process, the DTA asked for and was given the data on the 178. The DTA went through its records, and sent a response claiming that fewer than half were really dead, and of those, about a third had since been caught by the DTA itself. Most of the not-dead ones, the department said, merely appeared to be dead because the wrong Social Security number was in the system.

All of which, as I see it, simply proves the auditor’s point about the weak controls that were in place, but the administration points to it as damning counter-evidence. Patrick himself was quoted using those figures this week to assail the integrity of Bump’s report, saying that “a small fraction [of the 178 reviewed] are problematic.” Well, define problematic, bub.

You can see the disconnect here. Bump’s side thinks it’s talking about what it found in April 2012, when it ran that particular test, and when the DTA was kind of a mess. Patrick’s side thinks the public is hearing that the DTA is a mess now, and is essentially trying to dispute that notion by assailing the data. Which is annoying to the auditors, because that’s not what they ever said the data showed. But, maybe if the auditor had made that clearer we wouldn’t have that problem.

Another contributing problem, at least as I gather from the auditor’s side, is that the DTA wouldn’t show its work on the 178. That is, the auditors asked for the materials substantiating the claims that, for example, “66 of [the cases confirmed as non-deceased] were a result of data entry error of the recipient’s SSN into the Department system.” The auditors wanted the materials, so they could properly reply to this response in the report. The DTA, they say, never provided it.

Meanwhile, the auditors had run their next check, in September 2012, which found 1164 such dead-SSN matches in the main benefit-recipient database. The DTA asked for those records too, but this time the auditor refused — it seems they were disinclined to give the DTA more data to dispute, until the DTA handed over the materials disputing the 178. Whether that’s how it really went down or not I don’t know, but the impasse has held to today — and on Friday I was still getting conflicting information  about whether an agreement had been reached about turning over the data on the 1164.

Monahan repeatedly insisted to me that she wants the records only in order to pursue any instances of impropriety, as she would with any tip about fraud. That doesn’t strike me as her entire motivation. It certainly isn’t Polanowicz’s stated motivation in his letter Thursday, in which he wrote that he is renewing his request for the 1164 “in light of the inaccuracies found with the guardianship cases you provided DTA” — in other words, so we can prove you wrong like we did with the 178. Patrick said essentially the same thing when he spoke publicly about those records the same day.

I spent a lot of time frankly baffled at why the DTA needed the 1164 from the Auditor at all; it seemed to me that since those matches had been found by checking the DTA’s own database against a SSA database the DTA uses, they should be able to find the records themselves. I think I’ve got a handle now on how the system works and why it’s not that simple, which I won’t bore you with. The simple version is that the DTA’s system syncs with the SSA’s nightly, and updates on both sides are checked, with potential problems (like a dead person’s SSN) raising an alert for a DTA case worker. But after that sync happens once, the flag won’t get raised again — so if the case worker never actually resolved the discrepancy, it could just stay in the system indefinitely without notice.

It’s not hard to imagine how 1164 of those legacy problems could build up in an understaffed, overburdened, and poorly managed office, which the DTA had been up to the time of the audit. And it’s legitimate for the audit report to show that. I can also see why the administration would want to take those records and demonstrate to the public that there aren’t really 1164 benefit-collecting dead people out there.

But by pursuing this line publicly — that the data is somehow wrong — it seems to me the Patrick administration has done itself a terrible disservice. They are helping to drive the incorrect perception that the audit is describing current conditions, rather than emphasizing the improvements that have been, and continue to be made. (As one data point on this, all of the panelists on Friday’s WGBH Beat the Press seemed to have the former impression, with none seemingly aware of the latter.)

Why would the Patrick administration do this? This is purely my own speculation, but: JudyAnn Bigby.

I think Patrick, and others in the administration, are particularly sensitive about maintaining Bigby’s reputation; I would argue that they did a good job ushering out the former EOHHS Secretary with a minimal amount of “in disgrace” coverage, and brought in experienced administrator Polanowicz with minimal “to clean up her mess” coverage.

A common perception of Bigby, which I share, is that she was in many ways a terrific EOHHS Secretary, with the vision and agenda-setting skills needed, both to refocus the departments after years of Republican neglect and to implement the far-reaching new health care coverage system — but that she was also a poor organizational manager, in way over her head running that massive bureaucracy.

After six years, and tremendous accomplishments, the downside arguably caught up, most notably in the drug lab tampering and the compounding pharmacy scandals. DTA would make three, and three’s a media-friendly trend, right?

So, while I doubt they care much about throwing Curley further under any buses, I suspect Patrick and the administration really don’t want this DTA report to become part of the “bad Bigby” narrative. And that, I suspect, is a big part of why they are trying so hard to fight the audit that, as I started out saying, nobody really disagrees with. And by fighting, they are prolonging the story and making it worse. They need to let it go.