Now that the House has officially rejected the Senate’s budget bill, the two chambers soon will begin conference negotiations to set a final budget that likely will be subject to a veto by Gov. Roy Cooper. A number of practical and political factors will influence spending decisions during this process.

We add another perspective with an approach we call “reverse logrolling” or a “consensus-based” budget. Traditional logrolling takes spending that one person or group wants and adds it to the spending that another person or group wants. Higher overall spending is the result. Reverse logrolling 1) drops the items that would increase spending in either budget, 2) takes the lower spending when an item is in common but the amount is not, and 3) accepts cuts wherever they occur.

For Fiscal Year (FY) 2019-20, this yields a budget of just $23.5 billion, $432 million less than either the House or the Senate appropriated and just $288 million more than the base budget. Using the same methodology, the State Capital and Infrastructure Fund would provide $721 million in debt service and $348 million for capital projects.


At first blush, the House and Senate budgets look nearly identical. Each chamber would spend $23.9 billion for current operating expenditures and $721 million for debt service on past capital projects. Each reduces taxes on individuals and businesses, dedicates new money to keep juvenile offenders out of the adult criminal justice system, provides pay raises for state employees and teachers, funds several capital projects including major information technology initiatives, and implements Medicaid transformation without expansion.

Look deeper, however, and the differences are not simply that the Senate dedicates an additional $699 million in cash balance and tax revenues to savings, capital, hurricane recovery, and unreserved cash balances. In fact, just $18 million in net new spending ($83 million in new recurring expenditures and $65 million in one-time cuts) is common between the two budgets.

The biggest difference is the ways that budget writers chose to compensate state employees. The House budget dedicates $150 million of its $202 million in pay raises to teachers and other school-related employees, but they receive just $97 million ($32 million in one-time bonuses) of the Senate’s $215 million in raises. In addition, the House delays its compensation increases, which reduces spending in the first year of the biennium and more the second year. In other areas as well, the House spends less in the first year and more in the second year.


Although the Senate includes $139 million more in the State Capital and Infrastructure Fund (SCIF) than the House, both chambers agree fully on debt service, 12 capital projects, and assistance with water projects and rural broadband that would use $811 million from the SCIF. They agree to set aside funds for repairs and renovations and initiate seven other projects, though they differ in cost and timing. The Senate would direct all of the extra funds available plus another $50 million ($189 million total) to school and community college construction projects that the House did not address in its budget and Gov. Roy Cooper would have funded with a portion of his $3.9 billion bond proposal.

Applying the reverse logrolling model to take the lowest amount for common capital projects (and ignore those not in common) would yield projects totaling $1.07 billion, $120 million less than the House’s proposed $1.19 billion in spending and $260 million less than the Senate’s $1.33 billion. The big questions in this area are whether to spend or save the $139 million in new availability.


We do not expect budget negotiators to adopt this approach in full, but it does provide an illuminating sense of what priorities the House and Senate share and where their differences can be reconciled. It is worth noting that the biggest net savings are one-time reductions in Medicaid and film production grants, not permanent repeals of programs. It simply may be that the remaining programs are essential, effective, and efficient, though there are reasons to question that notion. There are nearly 300 items that only one chamber would have funded and thus are excluded from the reverse logrolling budget. Most of the remaining parts of the budget have room for lower spending in one chamber compared to the other.


DOWNLOAD FULL DATASET: Consensus Based Budgeting