Three Observations on the House Accountability Proposal

Was this forwarded to you by a friend? Sign up, and get your own copy of the news that matters sent to your inbox every week. Sign up for the On EdTech newsletter. Interested in additional analysis? Try with our 30-day free trial and Upgrade to the On EdTech+ newsletter.

In Friday’s post, I compared three different institutional accountability approaches in terms of the conceptual design and the key metrics used. The Democrats’ Financial Value Transparency & Gainful Employment (FVT & GE) regulations that are active, the Republican House Risk-Sharing reconciliation bill proposal, and the Republican Senate Gainful Employment for All reconciliation bill proposal. This week I plan to publish ~3 additional analysis posts on the two reconciliation bill proposals, given the importance of the next 2 - 6 weeks of negotiations that will likely determine how institutions are held at least partially accountable for student debt and non-repayment levels for the next half decade or more.

In this post, I’d like to share three observations that derive from data analysis on the House Risk-Sharing approach, which would require colleges and universities to reimburse a portion of the student loan debt in non-repayment status. These reimbursements would be offset by PROMISE grants that are based on formulas (non-discretionary and automatic) designed to financially encourage lowering of college costs and increasing of access and student completion. I have now improved the name-matching algorithms to allow visibility into <2-year institutions that award certificates but not degrees and to look at specific academic programs.

My thoughts after diving into the data:

1. There just isn’t much aggregate (federal) budget impact from risk-sharing, as only 1 - 2% of the proposed education savings come from institutional risk-sharing. This is a lot of change for just $3.5 - $6.2 billion in savings over ten years.

2. To a surprising degree, the Republican House seems to be following Democrat precedent and would transfer billions away from private nonprofit and for-profit institutions to public institutions.

3. Now that I can add the <2-year sectors (i.e. those offering nondegree programs only, mostly for-profit), you see similar patterns that reinforce this subsidy view that emerges from the data.

Let’s look a little deeper.

There Isn’t Much Aggregate (Federal) Budget Impact

Let me state up front, just because the aggregate budget impact is small, that doesn’t mean that the House Risk-Sharing proposal wouldn’t have a big impact. First, institutions individually would have big budget impacts, possibly losing or gaining tens of millions of dollars. The aggregation can hide that dynamic. Second, colleges and universities would face a direct financial impact on program net price, the level of student debt non-payment, and the earnings of both program completers and non-completers.

But despite this change and all the angst about the budget impacts, it is interesting that the aggregate federal impact is quite modest. Preston Cooper from AEI (conservative policy group) has done his own calculations on the public data to discern aggregate impacts, with a recent post noting the levels.

Cooper estimates total ten-year savings, whereas the House Committee on Education & Workforce (CEW) and the Congressional Budget Office estimate $6.2 billion savings from risk-sharing out of the roughly $351 billion in total CEW savings over the next ten years. This means that risk-sharing represents just 1 - 2% of aggregate budget impact.

Total disbursements of federal student loans plus Pell Grants is $152 billion per year, and the peak budget impact year in Cooper’s analysis has a net savings (reimbursement totals minus offsetting PROMISE grants) of $600 million, or 4% of the total.

Despite all of the budget savings rhetoric of Virginia Foxx last year and the House CEW this year, it doesn’t seem like aggregate federal budget is actually the driver.

Reimbursements from Private and For-profit Institutions Would Subsidize Many Public Institutions

Per Cooper’s ten-year estimates that include degree-granting and nondegree institutions, public colleges and universities would benefit by $7.7 billion (grants - reimbursements), whereas private nonprofits would lose $6 billion and for-profits would lose $3.9 billion. Special focus institutions would lose an additional $1.3 billion.

Per the CEW year-one estimates, public control institutions would gain $1.5 billion, private nonprofits would lose $884 million, and for-profits would lose $555 million.

If we break down the control into individual sectors (per IPEDS), including both degree-granting and nondegree institutions, we can see additional aggregate impacts.

If I showed you this chart out of context, you could be forgiven is you assumed this was a Democrat proposal.

Nondegree Institutions (aka <2-year sectors) Show Similar Patterns to the Degree-granting Institutions

One way to view the institutional impact is Net Impact per Student (grants minus reimbursements divided by number of students) vs. Program Net Price (tuition + fees + expenses, excluding grants).

Degree-granting view

In my analysis last week I focused on the three 4-year and above sectors and the Public 2-year sector (mostly community colleges).

Here you see the biggest beneficiaries (above the $0 line) as Public 2-year and Public 4-year sectors, and Private nonprofit 4-year and For-profit 4-year losing the most.

<2-year Nondegree view

Since so much of the FVT & GE regulations were based on nondegree institutions, particularly for-profit <2-year sector, I can now add this data. The view below is for public <2-year, private nonprofit <2-year, and for-profit <2-year sectors.

There are far more for-profit schools in this grouping, but a similar pattern holds in terms of Public <2-year institutions gaining and for-profit <2-year institutions losing.

Coming Up . . . Program Analysis

Since the reimbursement side of the House Risk-Sharing proposal is calculated per program and then aggregated to the institution, the next post will start to look at the CEW program data. A key question I have been prompted to look into is the relationship between Net Impact and individual metrics like tuition, net price, completion rates, etc. That is coming next.

The main On EdTech newsletter is free to share in part or in whole. All we ask is attribution.

Thanks for being a subscriber.