Some Thoughts on the Latest Round of Gainful Employment Negotiatied Rulemaking Part 1

more+
less-

I will apologize upfront for the length of this post.  Or that there will be multiple posts on this.  There’s just a lot to say after this latest session of gainful employment negotiated rulemaking.

For those of you that have missed the news, the Department of Education requested, and the negotiation committee agreed, to hold one more round of negotiations.  The sessions will last one or two days and happen somewhere between December 9-20.  If the goal is to achieve consensus around the proposal – or at least reach some agreement on some ideas – this is a good move.  I thought the Department made a mistake in not planning on holding three sessions in the first place.  While I understood the rationale – the Department has, after all, already held three sessions on this rule in the past, so everyone knows the issues – I thought not holding the last session failed to respect the normal process: the first session negotiators meet each other and talk philosophy; the second session they discuss actual proposals and coalesce; and the third session they negotiate over actual language.  Although many negotiators had a fairly solid grounding in the issues this time around, the process still remained the same, and wasn’t appreciably advanced by the Department’s release of proposed rule language prior to the first session.  I would, however, be remiss in not acknowledging that the Department’s dramatic revision to that first proposal prior to the second session also likely played a role in the failure to complete in two sessions.  By adding two new metrics, making schools comply with three of four of the proposed metrics, and not having data regarding how the two metrics would affect school programs, it may have been too much to expect negotiators to be in a position to vote on consensus this round.  Nonetheless, the Department deserves kudos for recognizing all of this and proposing to hold another session.

That said, I am curious how this final session will go.  Ostensibly, the purpose was to provide data to the negotiators on the effect of the new metrics.  What is unclear to me is whether the data will make any difference to the Department.  Indeed, the Department seemed to suggest that the main policy proposal (the four metrics) wasn’t going to change.  While we don’t have a transcript of the proceedings (the committee voted not to transcribe the proceedings on the first day of the first session), Ben Miller of the New America Foundation has done a live blogging of the sessions that squares with my recollection of the discussion:

Jones from Strayer says he appreciates the attempt to get data. He asks about the process going forward. He notes that some of the ideas the Department has put forward lies in the impact of the data. He asks if the terms of the rule appear to be set, what is the point of the data if there isn’t openness to using that data to find some critical underlying points. He says if the Department is going to have the data would it be used to revisit core components of the rule. Kolotos says it will give the data and we can discuss the metrics, but it must be done in one day and not have session after session. Jones says he in particular is concerned about the impact of complying with one of three metrics versus having to comply with all three metrics.

. . . .

Kolotos responds to Jones. He says he believes the Department put forward the right policy. The data should inform the policy, but it should not drive it. That seems to suggest that the policy should not be driven by the outcomes estimates but by what makes sense from a policy standpoint. (Italics in original).

I see this as an overreaction – perhaps an understandable one – to the court’s decision dealing with the last gainful employment rule.  As you may recall, the court in that case upheld the debt-to-income measures as the product of reasoned rulemaking but struck down the loan repayment metric.  While the court found that the Department based the debt metrics on “expert studies and industry practice,” which resulted in a “rational connection to the facts found tand the choice made,” the loan repayment metric was said to be based on an outcomes driven analysis:

The debt repayment standard, by contrast, was not based upon any facts at all. No expert study or industry standard suggested that the rate selected by the Department would appropriately measure whether a particular program adequately prepared its students. Instead, the Department simply explained that the chosen rate would identify the worst-performing quarter of programs. Why the bottom quarter? Because failing fewer programs would suggest that the test was not “meaningful” while failing more would make for too large a “subset of programs that could potentially lose eligibility.”  That this explanation could be used to justify any rate at all demonstrates its arbitrariness. If the Department had chosen to disqualify the bottom ten percent of programs, or the bottom half, it would have offered the same rationale: the rate chosen disqualified the percentage of programs that it was intended to disqualify, and to have disqualified fewer would have made the test too lenient while disqualifying more would have made the requirement too stringent. This is not reasoned decisionmaking. “As an expert agency, [the Department’s] job is to make rational and informed decisions on the record before it in order to achieve the principles set by Congress. Merely . . . picking a compromise figure is not rational decisionmaking.” In setting the debt repayment rate, the Department picked a palatable figure. Because the Department has not provided a reasonable explanation of that figure, the court must conclude that it was chosen arbitrarily.  (Decision, at page 31 (citations omitted).

As I read this case, the Department need not ignore what the data shows in coming up with the rule – the court only held that the data can’t be the sole basis for a rule.  Indeed, if a given rule would eliminate all or substantially all of a certain type of program, that should be a good reason to go back to the drawing board.  Now, I may be reading the Department incorrectly here; the folks I know at the Department are reasonable and would want that data so they could see the effect of the rule before they publish the rule.  I just hope the court’s decision hasn’t gotten the Department thinking otherwise.