All Articles
As part of building the state budget each year, the Legislature and Governor must make an assumption about how much revenue the state will collect. Because no one knows how much revenue the state will collect next year, leaders must rely on revenue forecasts. Both our office and the Department of Finance (DOF) provide periodic revenue forecasts that can be used for this purpose. These forecasts use the best available data to provide informed estimates of future revenue collections. Although they have limitations, they are important to the state budget process because they are a guide of which direction state finances are trending and offer an objective foundation on which the state budget can be built. In this post, we offer guidelines to help make the best use of these revenue forecasts—that is, to help them focus on the right questions, avoid overreactions, and be better positioned for the unexpected.
- Have Realistic Expectations About Accuracy. Decision makers should expect state revenue collections to differ from budget forecasts by billions of dollars in most years. Over the last four decades, the budget act revenue forecast for the budget year has been off by $7 billion to $10 billion in a typical year (in today’s dollars). Misses of $25 billion or more are not uncommon. These errors arise because revenues depend on factors that are inherently unpredictable, such as the economy, financial markets, and taxpayers’ decisions. These underlying factors are immensely complex and history offers us only sparse clues about how they work. As a result, revenue forecast errors are an unavoidable feature of the state budget. They also are ubiquitous. Every state in the country experiences meaningful forecast errors, often of a similar magnitude to California.
- Do Not Focus Too Much on Smaller Differences. Our office’s forecast offers a reasonableness check for DOF’s forecast. When DOF’s forecast for the budget window is within $10 billion of ours at May Revision (or $15 billion at the January budget release), decision makers should tend to view it as reasonable. Only when our forecasts differ by more than this would we suggest decision makers consider raising questions about the reasonableness of the administration’s forecast. To arrive at this conclusion, we estimated the past performance of a large variety of forecast models. No single model stands out as the best, but we can narrow in on a collection of high-performing models. Even among these top models, forecasts can differ by as much as $10 billion per year. This suggests that differences of this size between our office and DOF probably shouldn’t be taken as a sign that one of the forecasts is unreasonable.
- Weigh the Risks. In addition to selecting reasonable revenue assumptions around which to build a budget, decision makers also should try to weigh the chances that revenues end up higher than assumed against the chances they end up lower. One way to weigh upside versus downside is to compare our office’s forecast to DOF’s. While comparing these forecasts to determine which one is “right” may not always be fruitful, the difference between them can still offer valuable insights into the balance of risks. Historically, when our forecast is above DOF’s, actual revenues tend to come in somewhat above DOF. The reverse also is true. Whether there is more upside or downside can inform key budgeting decisions. For example, if the state is facing a deficit and more downside risk, it could make sense to make contingency plans, such as automatic mid-year spending reductions or revenue increases. If facing a deficit but with more upside in revenues, automatic mid-year spending augmentations could be warranted. In contrast, if there is a surplus but more downside, decision makers may want to be more cautious about making new, ongoing commitments.
- Interpret Real-Time Tax Collections with Caution. Real-time tax collections provide an important means of tracking and reevaluating revenue forecasts throughout the year, but decision makers should do so with caution. Monthly tax collections are noisy. For example, state and federal policy changes frequently distort how and when tax payments are made. Similarly, the timing of holidays and key collection days can result in collections coming in one month this year and another month the next. Further, some important drivers of tax collections, such as financial market conditions, can change dramatically throughout the year. This means patterns early in the year are not guaranteed to persist. For instance, historically, assuming tax collection patterns in the first half of the fiscal year persist for the full year tends to miss the mark as a prediction of actual full year collections by $10 billion in a typical year (in today’s dollars). For these reasons, decision makers should be careful not to read too much into short-term patterns. Comparisons of monthly collections to budget forecasts, even when looking over multiple months, are best used as a gauge of whether revenues are, in general, improving or deteriorating.
- Do Not Equate Revenues and the Economy. When building the state budget, decision makers likely will want to understand the current state of California’s economy—that is, how the state’s workers, consumers, and businesses are doing. To do so, decision makers understandably may look to the revenue outlook as a measure of economic strength. We suggest, however, they be careful about doing so. Revenues are shaped by many factors, such as financial market conditions, the state’s tax structure, and taxpayer behavior, that often have a tenuous connection to the broader economy. As a result, revenue trends can diverge considerably from trends in the state’s economy. For example, in 2020-21, the state’s unemployment rate was double what it was in 2018-19 and, yet, state revenues were nearly 50 percent higher. In contrast, the state had solid employment growth and falling unemployment in 2016-17, but tax revenues grew at only half the historical norm. Such disconnects make it difficult to tell a clear and simple story about the link between the economy and state revenue.