This is the ninth in a 10-part series outlining New Jersey’s fiscal fundamentals. The goal is to demystify some of the state’s financial challenges, and put them in context of the broader issues New Jersey faces. This series is also intended as a way to underscore the importance of state government in a year that will see a new governor and a new Legislature chosen by voters. Follow this link to see the.
In the preceding eight articles we discussed the principal aspects of the finances of New Jersey. It is also important to understand how the entire budget process works — as it is the principal means for developing and presenting the governor’s annual budget and having it approved by the legislature.
The agency charged with planning, developing, and overseeing the execution of the annual budget is the Office of Management and budget (OMB)—the office I had the privilege of leading for five years.
So, who better to tell us than the professionals who work in OMB. For the most part, I have taken the material that OMB publishes on its website on how the budget process works, and have made certain changes and additions, including adding a section on revenue estimating.
The diagram below is an overview of the five principal parts of the budget process — from planning, to finalizing the governor’s recommendations, to working with the Legislature, to developing and finalizing revenue estimates, to working with the governor to review the final Appropriations Act the Legislature returns to the governor, to the final signing by the governor.
The state budget process is designed to result in budget decisions that are informed by performance, with a focus on furthering agency core missions. Initiation of the budget process begins during the month of August, some 11 months prior to the start of the year for which the budget will be effective. The New Jersey state budget cycle is set on a fiscal year basis, which extends from July 1 to June 30 of the following year.
To formally initiate the process, the OMB provides technical budget instructions to the departments, as well as preliminary budget targets in September. This enables the agencies to determine how to best allocate available resources to meet their core missions in the coming budget year. The ensuing planning process includes reviews of the governor’s program priorities, as well as economic forecasts, demand assumptions, and analyses of selective program areas.
Agencies prepare planning documents that describe: 1) their ability to achieve their core missions at the preliminary budget level ; 2) the agencies’ priorities for reduction of current services if requested, and the impact of such reductions on their core mission areas; and 3) priority packages representing either expansion of current programs or new programs. OMB reviews the planning documents with the agencies from November through mid-January, when preliminary recommendations are agreed upon.
During the months of January and February, the director of OMB reviews budget recommendations with the state treasurer, the governor, and the governor’s staff. The governor makes the final decisions in early February.
The planning portion of the budget process usually culminates in February with the submission of the governor’s budget message, representing the governor’s recommendations to the Legislature for allocating resources. The budget message is delivered to the Legislature on or before the fourth Tuesday in February (unless legislation is passed to extend the date). The budget is the single most important policy statement that the governor makes as he or she allocates the state’s resources for programs and services.
The annual review process for capital spending requests and recommendations that are principally funded by bond revenue, has several steps, and runs somewhat parallel to the process described above. All state departments requesting capital funding must submit a seven-year Capital Improvement Plan to the New Jersey Commission on Capital Budgeting and Planning. Each capital project request must include an operating impact statement, which outlines the carrying costs, personnel costs, and other expenditures that impact the operating budget. The commission schedules public hearings, analyzes the capital requests, and recommends projects to the governor. The governor, in turn, selects projects to be recommended in the annual budget message.
The Legislature, through a series of hearings conducted by its appropriations committees, reviews the governor’s budget and makes changes. The legislature also reviews the revenue estimates included in the governor’s budget and, based upon several additional months of actual revenue collections in the current fiscal year, makes adjustments to these revenue projections and surplus estimates.
The budget, including changes made by the legislative committees, then must be approved by the Senate and the Assembly. According to the New Jersey Constitution, a balanced budget must be approved as an Appropriations Act and signed by the governor before July 1. After the Legislature passes the Appropriations Act, the bill is sent to the governor.
The governor may sign the bill, conditionally veto it (return it for changes), or veto it absolutely. The governor also has the power to veto specific appropriations (line items) or appropriation-language segments, some of which may have been added by the Legislature as a result of its review. The line-item veto allows the governor to reshape the final budget and ensure that appropriations do not exceed the certified level of revenues. And although the line-item veto allows a final budget that reflects the governor’s priorities, budget items cannot be added at this stage to balance them. The veto can also be overridden by a three-quarters vote of the Senate.
As part of the final Appropriations Act, the governor must “certify” the level of revenues in order to meet the constitutional requirement of a balanced budget. The final approved budget, which includes the governor’s line-item vetoes and certification of revenues, is the Appropriations Act and becomes an effective tool for fiscal control and for monitoring program effectiveness.
Throughout the course of the fiscal year, the Legislature has the authority to pass legislation that provides funding for programs above and beyond those provided for in the Appropriations Act. The additional amounts of funding are referred to as “supplemental appropriations.” The director of the Office of Management and Budget also has statutory authority to approve certain supplemental appropriations at any time during the fiscal year by virtue of authorizing budget language contained in the Appropriations Act. These changes would be routine and not of a policy nature. This is accomplished and documented by the issuance of Directory Letters by OMB.
When the Appropriations Act is finally approved the OMB loads all the appropriation data into the state’s core accounting system. All fiscal transactions are recorded in the accounting system so that at any point in time the OMB has the ability to oversee the entire fiscal transactions of the state by individual line items. No expenditures can be made except through the accounting system so as to ensure that proper checks are made to determine that the proposed expenditure is valid and is paid against valid appropriation authority.
At the end of the fiscal year (June 30) the OMB is also charged with preparing the state’s annual financial statements for all transactions and accounts, including all revenue received. The annual financial statements are than audited by the state auditor, who is appointed by the Legislature. The audit must be performed by an entity independent of the Executive branch.
Estimating the amount of tax money the state will collect is one of the biggest challenges in the entire budget process. In a normal process the estimators (usually trained economists) consider and analyze the U.S. macro economy; the state’s economy — current and projected for the next 12-18 months; and historical and current tax collections. All of these elements will help the estimators arrive at a number for each of the major taxes — the most critical ones being income and sales, and to some degree, the corporate income tax.
Critics often argue that the estimating process would be improved if a multiyear outlook were adopted or more outside economic experts were involved. While any given forecast covers a 12-18 month period, the revenue forecast is formally updated every six months and closely monitored each month. In this way, changes in economic conditions or revenue collection patterns can be quickly identified and brought to the attention of policymakers.
Extending the forecast window does not add much to accuracy of the forecast for the current or next budget years given the uncertainties of anticipating the future. Likewise, external advisors, while they can be useful at the macro level, rarely have the time or experience to understand the details and meaning of the monthly revenue flow. These flows like sales tax collections and monthly wage withholdings are the most immediate signals of underlying economic changes that may appear only months later in the data.
Having said that, it is still beneficial for the Executive branch to project five-year revenue and spending projections if only to develop a broad view of what future budgets might look like under various tax and spending projections — multiple scenarios would be desirable, and if the Executive is really brave, the projections should be made available to the Legislature and the public.
A major observation — no estimating process is foolproof. Consider if the estimators do everything right and perform all the sophisticated analytical processes available to them and they are off by 2 percent. Well, a 2 percent error on a $35 billion tax base is $700 million. Or just think simply of the income tax estimate –a 2 percent error is $300 million. Or, let’s say in the middle of the fiscal year the economy takes a significant and unexpected downturn — the Great Recession of 2008-2009, for example —one could be off by 5 percent. The best estimators in the world would be in error — big time. Obviously, it’s necessary to maintain a healthy surplus and a rainy day fund.
But the process is important. One wants the revenue estimating to be as apolitical as possible. There are basically two ways states do revenue estimating: a single-point estimate or a consensus estimate. New Jersey follows a single-point estimate — to the extreme. At the end of the process only the governor certifies the revenue.
As noted above, the governor makes the initial estimate when the budget is delivered in February; the independent Office of Legislative Services (OLS) makes its estimate in March — and both refine their estimates in late May after April 15 income tax collections, as well as more data concerning actual collections of all revenue in the current year and updated data concerning the economy in the next 12 months. Both estimates could be close — or they could be far apart — as they have been in several years.
Sometimes revenue estimates are far apart not only for the current year but also for the projected next year. One can imagine what wide variance could lead to: unnecessary political wrangling between the Executive and the Legislature, particularly regarding what programs can be expanded, sustained, or reduced.
And when the governor has the final say regarding revenue estimating — guess who wins? New Jersey should do what many states do — use a consensus estimate process. There are many models to be followed. A simple one could be the formation of a group of experts and leadership from the Legislature and Executive. There could be as few as three or as many as five — not too many, or a consensus would never be reached. Ideally, the group would consist of an Executive agency expert (OMB director), OLS director and trained economists from Rutgers and Princeton. They would meet privately to deliberate about the data and then in public to discuss their final consensus recommendation for all tax revenue, especially the big three –income, sales, and corporate taxes. Their recommendation, not the governor’s certification, would be the determinant revenue estimate.
To paraphrase the National State Association of Budget Directors … Unfortunately no specific revenue-estimating process will yield the exact correct numbers, but consensus is preferred …
There are many risks when revenue estimating is not consensus driven, including risks of political gains by accusing the forecast of being biased. . I would argue that a consenus forecast would put the focus on spending and tax policy — not revenue disputes.