Improving Forecasting Techniques In Uncertain Economic Times

By Scott Pattison, Director, Tax Policy and Research, Federation of Tax Administrators

October 29, 2024

As we approach 2025, state governments are developing and passing budgets, not an easy task in a time when they confront a great deal of uncertainty about events outside their control that impact the economy and thus incoming revenues. There are major conflicts in eastern Europe, the Middle East and in Asia that can lead to disruptions in trade, supply chains and rising energy prices. A momentous presidential election in the U.S. is bringing about uncertainty as the new Congress and President determine major policy changes over the next few months.

For state and local governments, this uncertainty makes planning for their financial future particularly challenging. Fortunately, the economists and other experts seeking to forecast the future finances of state and local government have tools and data to rely upon that predicts state and local revenues over the coming months.

Good forecasting of the amount of revenue coming into state or local governments is a critically important function handled by state and local government economists and revenue experts. Revenue forecasts are essential for legislative and executive officials so that they have a good idea as to how much money they can spend for education, health care, infrastructure maintenance and a long list of other functions they perform. As government officials develop and adopt their budgets, the revenue forecast helps them determine if they need to cut spending or raise taxes or do both.

In addition to geopolitical factors, the U.S. economy has numerous risks including high interest rates, moderate inflation, and potential consumer spending constraints that can create recessionary pressures. An additional challenge for states is that the billions in funds provided by the federal government through Covid relief legislation will be ending. States will have to have obligated the federal money by the end of this calendar year 2024 and must spend it all by 2026.

States received a total of almost $250 billion from the March 2021 legislation. That’s a lot of money to be going away in the next year. In addition, the two federal bills in 2020 providing stimulus to the economy created economic benefits that helped state revenue grow at unprecedented rates of 16%. That’s way up from an average of closer to 4% year-over-year growth, but it, too, has a set life span and won’t be providing cash influxes forever.

Fortunately, forecasting government revenues has become a more efficient and potentially more accurate task thanks to advancements in data analytics and better and faster data collection. Artificial Intelligence (AI) is also getting a lot of attention and may over the next few years offer some benefits to forecasting.

Improvements in technology in recent years have made collecting economic and revenue data faster and also cheaper. Governments can now use less resources to gather and access data on economic activity and risks to the economy, such as various world conflicts and financial system challenges. Data and better information on business activity and consumer behavior can be used along with economic indicators. There are multiple sources available for all kinds of economic data. This data is used to forecast revenues based on economic conditions and factors unique to the particular state or locality.

Technology improvements allow governments to collect data faster, in some cases close to “real time”.  Getting data faster enables better forecasting since the revenue forecasters can make quicker adjustments to sales and income tax forecasts since they’re sensitive to quick economic changes. There are some good recent examples as to how quickly unexpected events occur and cause huge economic shifts. In March of 2020 when COVID spread caused much of the economy to shut down, states had to quickly adjust their revenue forecasts downward. When the economy began quickly recovering in a few months, revenues improved quickly and significantly. Adjustments to the forecast had to be down quickly so having data as soon as possible is critically important.

Finally, forecasters are utilizing simulations of possible economic trends and using advanced statistical techniques to come up with various possible scenarios to predict the impact of different occurrences on the economy and thus tax revenues. Having several scenarios help officials make better decisions. Knowing the most likely scenario but also other possible outcome should the economic conditions change. For 2025, state officials have to be prepared for a possible recession. If the current slowing economy is indeed a “soft landing”, revenues will not decline as significantly as they will if there is a recession. And a recession is certainly possible.

For forecasters, there also is a bit of an “art” portion to the science of predicting future government tax revenues. They use their expertise and experiences to adjust their forecasts. They carefully analyze macroeconomic data and examine historical data. All of this activity together assists them in giving decision making officials the best information possible. Finally, utilizing different scenarios helps to inform the process by showing what can happen depending on the particular circumstances. Predicting the future is close to impossible but fortunately government economists have increasingly better tools and techniques to provide the best forecasts they can.

The contents of this blog post reflect those of the authors, and not necessarily those of the GFRC.