Many people in Colorado are angry, frustrated or just confused after the U.S. Department of Education passed over Colorado while awarding grants in the Race to The Top (RTT).   After a similar loss in round one, Colorado made advances in the last legislative session that were politically painful to achieve, but repeatedly praised and defended because they would help Colorado’s chances at winning this competition.

Nevertheless, the state managed to rank only 17 out of the 19 finalists.  After the Feds decided to award only 10 grants, friends in Colorado are asking, “What’s up with that?”

A few assertions appeared throughout this process that deserve a response. Over the last year, I have repeatedly heard statements such as the following:

“But President Obama and Secretary Duncan want to support all our reforms, and they especially want to help Colorado’s new Senator, Michael Bennet.  Since he is such a reform champion, and potentially vulnerable in the next election, surely they’ll help him bring home the bacon to Colorado.”

Or

“This is too important a competition to leave it up to the bureaucrats and some peer reviewers. With 3.4 billion dollars at stake, the political appointees are definitely going to make sure the “right” states win, and we are certainly among the chosen in Colorado.  They can’t deny us!”

As these ideas, or more subtle versions of them, were raised over the last year and a half, I have politely tried to explain that federal grant competitions don’t work that way.  As a former employee at the U.S. Department, I worked around various national competitions for federal funds that were awarded through similar competitions – albeit for considerably smaller amounts and with lower profile programs and less political stakes involved.

I started out as neither a political appointee nor a career official at the department. I began serving with a temporary expert appointment and eventually I wormed my way into the Department’s career staff where I led the Department’s Public Charter Schools Program.

I’m not a lawyer so don’t ask me to quote the code, but all these competitions are run under the auspices of various federal rules and regulations and within the limitations of the statutory provisions for that particular grant program as they are written by Congress.  They are administered by career professionals rather than political appointees.  Peer reviewers rank the proposals in a “slate,” with the highest-scoring application at the top and the rest of the applicants listed in order of their score down the list.  The funds are awarded to each high ranked applicant in order until the available funds are used up, or until the remaining applicants were ranked so poorly that they were basically determined ineligible.

Keep this piece in mind if you ever apply for these grants. Unless it is in the criteria of the review, the amount applicants ask for is not really part of the decision-making process until after the ranking takes place.  Only then do the amounts matter, but only as they are totaled up in a running total as the officials designate grantees further and further down the slate.  When they use up all the available funds at the funding levels requested the applicants, a line is drawn through the slate. Those above get what they asked for, and the rest of the applicants are bitter losers.

Political appointees are not allowed to do much beyond provide input into the creation of the RFP and the rubrics, or the writing of selection criteria for peer reviewers. Even the rubrics are subject to the rule-making process, or a set of “generic” criteria are applied for low-profile programs.  Once those materials and the list of reviewers are set, the political folks would be violating the rules and regulations if they were making decisions.  If they put pressure on the non-political folks to change scores, or if they insist that the career administrators pass over one applicant and award funds to an applicant that was given a lower score, the political appointees (and their bosses) get in trouble.  It is juggling of this sort over Reading First that got folks in trouble during the last administration.

But there is one thing that political appointees could have done, or argued for, that might have changed the outcome after the ranked slate of applicants was determined by the application process.  They could have tried to argue that the awardees should get smaller grants, which frees up money to award grants to the next few applicants on the slate.  However, in this case, that is still extremely unlikely to have helped Colorado get funded. This is because of Colorado’s low ranking in the competition and the Department’s interest in ensuring that these grants be so incredibly big that states would do anything to get them.

An analysis of the states that were competing, their scores in the competition, their student populations, and the size of the awards granted to the lucky winners quickly shows us why any political shenanigans (even the quasi-legal effort to reduce award amounts to create money to fund lower-ranked applicants) were unlikely to have helped Colorado in this case.

The other forms of string-pulling that Colorado may have hoped, or other states feared, would generally produce scandals so distracting, and so potentially damaging to the Administration, that the risk of leaving Colorado – and even Candidate Bennet — without this prize would pale in comparison to the risk of trying to help them unscrupulously.

So if we assume adjusting amounts was the only strategy at their disposal, how could it have worked? It depends on how much states get and whether those amounts could have been adjusted enough to fund Colorado – not likely.  Figure 1. below lists awards received from the RTT in the first two competitions, the student populations of each state, and the award amount calculated on a per-student basis (as if every student in the state had a portion of the RTT grant award spent on their education.) In truth, districts are guaranteed only 50% of the total, and how districts spend the money is dictated by the application.  But the measure of each state’s student population gives a rough measure of the scale of their education system and the capacity to spend a huge grant like this.

Figure 1.

The RTT was not a formula. It was a competitive program. However, the amount a state could compete for was determined by a crude formula that set broad parameters for possible funds in each round of the competition based on relative state size. Under this formula Colorado was applying for $175 million.  All of the awards certainly pass the “ginormous test” by any state’s standards of avarice.  However, the variation in the dollars per student raise questions about the wisdom of such a crude formula driving the size of such huge awards.

There is unquestionably much that can be done to improve D.C. or Delaware’s education reform efforts, and maybe there are efficiencies of scale that make reform easier and cheaper in bigger states, like New York and Ohio — yeah, sure.  No matter how you cut it, it is hard to argue these tiny jurisdictions need four or five times as much money per student as Ohio (where reform is presumably a pretty cheap and easy affair).  But if you’re trying to fund additional states, these relatively excessive amounts for small states don’t help much because the totals would quickly be gobbled up by bigger states ranked higher than Colorado in the competition.

Given this distribution, and even despite regulations and announcements about the amount available to states of different sizes, there could have been room for political pressure.  A hypothetical discussion would go like this, “That seems like a lot of money for some of these states. Can they really spend that much? How about if you look at their budget and their proposal and see if there are areas where it is imprudent to give them as much as they asked?”

If the amounts requested don’t match the work proposed, or there was too much uncertainty about such budgeting issues, then the amounts funded would be negotiated by the career administrators and the states.  Again, the political appointees don’t get to do this directly.  But the political appointees could send serious signals to career staff, and given the scale of these proposals and the rush to get them in, inevitably some wiggle room in award amounts would be possible. The award amounts might plausibly then be reduced to see who else on the slate could be funded.

For arguments sake, let’s say that it was determined that the amount that Ohio actually received per student was a reasonable thing to expect of all recipients – and the original formula proposed in the competition was abandoned in favor of this “little adjustment” to match the Buckeye State.  Figure 2 below includes all states that were finalists in the second round, listed in a slate according to their rank.  The columns on the right indicate the amount the states would have received if total awards were run by formula, with a per student amount set at the minimum of the Round 2 awards ($229 per student).  The far right column indicates the total amount of spending given that per student amount.  As we see, if this strategy were pursued, three more states could have been funded within before the total amount available to spend expended was used up (New Jersey, Arizona and Louisiana).

Figure 2.

Continuing in this fashion, if the total grant award were reduced to $200 per student, with the current ranking, South Carolina would be the only state added. At $150 per student, it would add Illinois.  It would take a reduction to awards based on a formula of around $135 per student before Colorado and California would receive grants given the current ranking among the slate of finalists.  At that point the current round of successful recipients would have had to receive a total of $1.67 billion less to free up money to get down the slate all the way to Colorado. That’s a financial and political loss to all the current winners of an average of 167 million dollars less than what they actually received in the competition.

If Colorado were ranked 11^th out of 19 finalists, then the kind political pressure that might reduce a few grants to fund one more could have come into play.  Entering the part of the process where political pressure is most likely to benefit a friend ranked 17^th out of 19 applicants made that help entirely unfeasible. Apparently New Jersey, the unfortunate state on the bubble, isn’t high enough on the Administration’s list of political priorities to incent them to play games with this competition.

Looking at a spread sheet like this I suspect any political shenanigans start to look mighty risky, with many more losers than winners.  Even for the unscrupulous, a risk-benefit analysis would likely let this competition stand as originally conducted.

While there may be many reasons why people in Colorado are right to be angry at being slighted by the results of this race, the problems come from the score the state received and our rank relative to the other competitors.  There is plenty to gripe about in the process that produced those scores — so there is no need to stop feeling indignant.  But the problems come in the creation of the RFP, the rubric (to the extent they used one) and the various reviewer’s vagaries of scoring. It doesn’t look like the Western states had a chance, or that reviewers understood semi-rural states with their mixed levels of union representation and local control, or the courage Colorado’s leaders took to enact the reforms that are necessary yet offensive to some unions.

That still leaves plenty of reasons to raise your blood pressure. But these are different than the political ones I’ve heard. At least Colorado’s failure is not due to some imagined failure of political leaders to pull the strings a few people wish they had.

Popularity: 28% [?]