by Greg Sobetski, Chief Economist, Legislative Council Staff
Well, it’s April again. All of your bills are either dead or awaiting their fate among a logjam in one of the Appropriations committees. You’re bracing for the long nights ahead and scrambling for a small pool of money available for your bill and too many others. Most of all, you wonder: why do we wait every year to conduct what feels like the great majority of legislating in the last three‑and‑a‑half weeks of a four-month session?
You probably know the answer. The Long Bill, the state’s budget bill, won’t pass until sometime in mid‑April, and most financial decisions wait until after the Long Bill when we know how much money is available for legislation. So maybe your question is: why does the Long Bill come so late?
There’s a lot going on there. The Long Bill is a product of months of Joint Budget Committee (JBC) hearings with its own staff, the Governor’s office, and executive agencies. But it also needs one key ingredient that isn’t available until around St. Patrick’s Day: the pivotal March revenue forecast.
The March forecast is one of four quarterly forecasts the Legislative Council Staff Economists produce each year, and the only one that we publish during the session. Our work on it begins in early February, but actual revenue forecasting doesn’t begin until March. That’s because we don’t have the data we need until then: we’re using cash fund revenue data through January, which is delivered at the end of February, and General Fund revenue data through February, which is delivered on the fifth business day of March.
Say we wanted to move the forecast earlier: that’s possible (with legislation), but it presents some tradeoffs. Given the amount of work required to transform raw revenue data into a forecast, it’s not feasible to move up the forecast and still use February General Fund revenue data. The February report is revealing, in that it’s the first report that shows income tax return data from spring filers. But it’s also fickle, in that people who file their taxes in February aren’t a representative sample of the full filer population, and we have to be careful about what conclusions we draw from the February data.
The more interesting problem is what happens if revenue deviates mid-session from an earlier forecast used by the JBC for balancing. Let’s say that the JBC balanced the budget to a February forecast that used no spring income tax filing data. Then, during the session, but after the Long Bill passed, new data became available and showed the forecast was wrong (it happens!). What then? Would the JBC re-open the budget to make new cuts? Even the possibility of this scenario might cause legislators to wait until late in the session to move bills through Appropriations. Both passing bills for which there’s not enough money, and PI’ing (not passing) bills for which money becomes available later, are uncomfortable outcomes for members.
And it’s not just the revenue picture that changes. This year, our February conclusions about the economy were upended when the United States and Israel launched a joint attack against Iran at the end of that month. We needed to rebuild our economic expectations to account for upgrades to our forecasts for oil prices and inflation. When more time elapses between forecasting and final decision making, it becomes more likely that some event will upend the assumptions that were used to build the forecast.
So, there you have it: the forecasts come out in mid-to-late March, the earliest it can be published while incorporating any amount of spring income tax return data. The Long Bill happens after that, and then there’s a lot of substantive policymaking to be done in the last few weeks of the session. Moving the forecast up is possible, but it would require legislation, would weaken the predictive value of the forecast, and could cause downstream legislative issues.
While it feels like we’re days late, we’re at least not dollars short.
