State Revenue Forecasts

Updated February 2008

Contents
Introduction
The Revenue Forecast Process
A Revenue Forecast Table
Forecasts and Results, 1999-2009
Keeping Track Month by Month

Introduction
Three times in every two years, Indiana forecasts revenues for future fiscal years. Forecasts are needed for the budget process. The General Assembly must know how much revenue will be available in order to decide how much to spend.

Unfortunately, predictions are never perfect. Revenues came in above projections throughout the second half of the 1990s.  Starting in fiscal 2000, revenues fell short of projections.  The shortfalls became huge in 2001, 2002 and 2003. This was the main source of Indiana's budget difficulties during the first half of the decade. Then, from 2005 to 2007, revenues exceeded projections again. The economy recovered. The outlook at the end of calendar 2007 was so rosy. The December 2007 revenue forecast reduced projected revenues for the 2008-09 biennium.

The Revenue Forecast Process
In every odd-numbered year, the Indiana General Assembly meets in a "long" session, to write a state budget for the next two fiscal years. A budget is a spending plan, a decision to authorize state agencies, departments and institutions to spend particular dollar amounts. Before the legislature can know how much to spend they must know how much they’ll have. That means estimating future revenues from state taxes.

Revenue predictions are made three times every two years. The first predictions for an upcoming biennium are usually made in December prior to a budget-writing year. In December 2006, for example, the state revised its prediction for the rest of fiscal 2007, and made its first predictions for 2008 and 2009. In April of a budget-writing year, just before the session ends, a revised forecast is made. This allows the legislature to include new information about actual revenue collections in December through March. In April 2007, forecasts were revised for what little remained of fiscal 2007, and for 2008 and 2009. Then, the following December, before the non-budget "short" session, forecasts for the current and next fiscal years are revised again. In December 2007, the state revised its prediction for fiscal 2008 (which started on July 1, 2007) and for fiscal 2009.

Indiana’s way of projecting revenues has one big advantage. Indiana makes a "consensus forecast." Representatives of the Governor’s office, both houses of the General Assembly and both parties all agree on the projections before they are announced. Without a consensus forecast, the Governor’s office might announce one set of projections, and the legislature another. Each political party could have its own, and lobbyists might chime in with projections too. Much of the session’s debate would be about which projections to use, not on the substance of legislation. Getting a consensus on revenues before the session begins eliminates this problem.

The projection process moves forward along two tracks. One group, called the Economic Forecast Committee, works up a prediction of what will happen in the Indiana economy. Currently this committee is composed of economists from Ball State University and Indiana University, and from private industry (there may be some turnover in personnel on this committee before December 2008). The comittee forecasts three economic indicators, U.S. gross domestic product, Indiana non-farm personal income, and the rate of inflation.

Meanwhile, another group called the Revenue Forecast Technical Committee works on models of Indiana revenues. This committee is composed of the fiscal aides to the House and Senate party caususes, an appointee of the governor (currenty an economist from Indiana University), and the chief revenue forecaster from the State Budget Agency. The revenue models are developed by the Budget Agency and by the Legislative Services Agency, with the aid of an economist from Purdue University (who also writes this webpage).

The models relate Indiana revenues to the economic indicators that the Forecast Committee will predict. Data from past years are used to measure the revenue relationships. The sales tax model, for example, is an equation that relates sales tax collections to Indiana personal income. The models change a bit every year, but in December 2007 each $100 increase in Indiana income was predicted to increase sales tax revenue by about $2.13. Models are developed for the sales tax, individual and corporate income taxes, and the cigarette, alcoholic beverage and riverboat wagering taxes.

Each model is reviewed and approved by the Technical Committee members. This is how a consensus forecast is achieved, since the committee's members represent both parties, both houses of the legislature, and the governor's office.

The Forecast and Techical Committees work independently. This is an attempt to make the process more objective, so that neither the Forecast Committee nor the Technical Committee can adjust its predictions and models with a particular revenue goal in mind. Just a day or two before the revenue forecasts will be announced, the Forecast Committee releases its economic predictions to the Technical Committee, and the predicted economic indicators are "plugged into" the revenue forecast models. The result is the revenue forecast, which is announced to the legislature and the public and is always well-covered by the news media.

A Revenue Forecast Table
The following table compares revenue forecasts in each year with the actual revenues collected.  Read across the rows to see how revenue forecasts for each year changed.  Read down the columns to see what revenues were expected as of the end of each fiscal each year.  Read down the diagonal (the numbers in italics) to see actual revenues in each year.

Consider the column (reading down) labeled 2005. The end of fiscal year 2005 was June 30, 2005. The entries in the 2005 column show what revenue was actually collected in fiscal '05, $11,436.5 million, and what revenues were projected for 2006 and 2007 as of that date. Since 2005 was a budget-writing long session of the legislature, these are the revised projections from April 2005. Looking down each column, then, provides a snapshot of what Indiana's revenue forecasters thought would happen, as of the end of each fiscal year. As of June 30, 2005, forecasters thought that Indiana would collect $11,757.4 million in fiscal 2006, and $12,378.3 million in fiscal 2007. The 2005 legislature wrote the biennial budgets for fiscal years 2006 and 2007 based on these predictions.

Now consider the row (reading across) labeled 2007. The first number in the 2007 row is that $12,378.3 million figure. It's in the 2005 column, because that was the prediction that the state made during the 2005 budget session of how much revenue would be collected in 2007. The 2007 budget was written based on this prediction. Move to the right one column, and there's a different figure, $12,348.8 million. This was the revised revenue projection made in December 2006 for the 2007 fiscal year. Move right one more column and the figure is $12,626.2. That's the actual revenue that was collected in 2007 (that's why it's in italics). Revenues for fiscal 2007 came in $247.9 million more than originally forecast, as shown in the column labeled "Change in Projections First to Last." The figure in the "Pct. Change Actual Rev. (italics)" column shows how actual revenues changed from 2006--a 4.7% increase.

A look at the 2009 row shows the effect of the economic slowdown expected in 2008 and 2009. The December forecast reduced estimated growth in Indiana income from about 4.5% per year to about 3% per year. That cut the revenue forecast from $13,377.0 (the forecast from April 2007) to $13,111.9 (December's forecast), a $265.1 million drop. If this amount is collected in 2009, it will still be an increase over 2008, but it will be less than was expected when the budget was written in 2007. Spending plans for 2009 were made based on a revenue forecast that had an addition $265 million. That's how revenues can fall short even when the increase from year to year. When they actually decrease from year to year, there's serious trouble.

Forecasts and Results, 1999-2009
We can use this table to tell the story of Indiana's revenue difficulties during the recession, it's recovery during the expansion, and the troubles the state may face in the near future.

Consider the column labeled "1999", reading down.  These were the revenue forecasts that were used to make the 2000-2001 biennial budget (the budgets for fiscal years 2000 and 2001).  Actual revenue in 1999 was $8,883 million.  As of the end of fiscal 1999 (June 30, 1999), revenues for 2000 were predicted to be $9,302 million, and for 2001, $9,773 million.  Steady growth was expected. For years during the 1990s expansion, revenues are exceeded projections. The economy (or the stock market, at least) grew faster than expected. In 1999, revenues were forecast to rise 4.7% in 2000 and another 5.1% in 2001.

It didn't happen.  Hints of a problem appeared when the numbers for fiscal 2000 were added up in July of that year (this is called the "closeout").  Actual revenues were $9,143 million, $159 million less than forecast the year before.  Revenues increased only 2.9% from 1999 to 2000, not the projected 4.7%.  The revenue forecast for 2001 was revised downward by $130 million, to $9,642 million.

Then came serious trouble.  In fiscal 2001, only $9,052 million was collected, over $600 million less than the revised forecast from the year before.  All together, revenues for 2001 were $721 million less than the original forecast on which the budget was based.  Revenues had actually fallen from the year before, by one percent. 

The revenue forecast made in 2001 for the 2002-03 biennium assumed the worst was over.  Growth of around 5% a year was expected--$9,529 million in 2002 and $9,955 million in 2003.  The budget was written based on the expectation of this normal revenue growth. But again, 2002 was a revenue disaster.  Actual revenues fell $820 million short of the previous year's forecast.  Revenues had fallen for a second straight year, this time by 3.8%.  No one could remember a time when actual revenues had fallen two years in a row.  The forecast for 2003 was revised downward by $705 million.

The budget difficulties and the approaching property tax reassessment spurred the General Assembly to restructure taxes. Tax restructuring was completed in a special session in June 2002.  Sales, cigarette and riverboat gaming taxes were increased. 

The 2002 and 2003 columns of the table show the expected and actual effects of restructuring on revenues. The row labeled 2003(1) shows projections prior to restructuring, and the row labeled 2003(2) shows projections and actual revenues after restructuring. Prior to restructuring, revenue was forecast at $9,250 million, a figure that had been revised downward by $705 million from the previous year's forecast. After restructuring, $10,251 million was expected. Restructuring was predicted to increase revenues by slightly more than one billion dollars in 2003. In fact, as shown in the 2003 column, revenues fell short of this projection by $371 million. Actual collections were $9,880 million. Restructuring certainly increased revenues--actual revenues jumped 13.4% in 2003--but not as much as expected.

Projections for 2004 and 2005 were made in 2003 (read the column labeled 2003). Again, significant increases in revenues were expected. The especially big increase expected between 2003 and 2004 was due to restructuring. The tax increases would be in place for the full fiscal year in 2004. But a substantial increase was expected in 2005 as well.

This time, the shortfall was small. Actual revenues fell short of the prediction by only $73 million in 2004. Despite this small shortfall, the revenue prediction for 2005 was revised downward by $190 million.

At the July 2005 closeout the Budget Agency reported actual revenues for 2005 at $11,437 million, $244 million more than originally predicted in 2003; $435 million more than predicted in 2004. For the first time since 1999 actual revenues exceeded predictions. Revenues had grown 7.7% over 2004. During the budget session of 2005, revenues for 2006 and 2007 were predicted to grow by 2.8% and 5.3%, respectively. The state did better than that in both years. At the closeout in July 2006, the budget agency reported that revenues were $303 million above projections. The next year, revenues exceeded projections by $248 million.

Indiana state revenues turned the corner in 2005. But the years 2000 through 2004 were a revenue disaster. In two of the years, 2001 and 2002, revenues had actually fallen from one year to the next. More important for budgeting, revenues had fallen $2.8 billion below projections during this time. The General Assembly uses these predictions to set appropriations for school aid, universities, Medicaid, property tax relief, corrections and all the other functions of state government. When they fall short, the state must use its savings. If savings run short, the state may scramble to find accounting tricks to balance the budget. With a revenue shortfall this big, though, Indiana had no choice but to cut spending growth, and raise taxes.

The opposite is true when revenues exceed projections, as they have these past three fiscal years, 2005-07. The state can pay for its planned spending. rebuild its fund balances, and sometimes, appropriate more money for added services in non-budget years.

The history of the 2000-2004 shortfalls made the news in December 2007 sound ominous. The economy was projected to slow in 2008 and 2009, and the December forecast cut the 2009 forecast by $265 million. Is this a mere one-year forecasting error? If so, Indiana has the fund balances needed to cover such a shortfall without significant budget cuts, and without tax increases.

Or is 2009 like the year 2000, when revenues first fell short? The year after that the effects of the 2001 recession and stock market crash hit state revenues, and the big shortfalls began. If that happens, Indiana will be in for several more years of tight budgets, fiscal gimmicks, and possibly tax increases.

Keeping Track Month By Month
Indiana's State Budget Agency publishes a report each month that compares actual revenue collections to projected collections. Monthly projections are derived from the annual projections based on historic monthly patterns of revenue collection. The reports show the difference between actual collections and projections each month. A total shows how collections compare to projections so far during the fiscal year.

Readers should know of a quirk in these reports. June's final report for the 2007 fiscal year shows revenue collections for July 2006 through March 2007 exactly on target. There is no difference between projections and actual collections. But go back to the March 2007 report, and you'll see the December through March revenue collections falling short of projections by $10.6 million. This is because the April 2007 forecast revision took actual collections from July through March as given. Earlier monthly forecasting errors are zeroed out. Likewise, July through November forecast errors were zeroed out by the December 2006 forecast revision. It's important to check the heading of the comparison tables to see which forecast is being used for the comparisons.

Still, the monthly comparisons are useful. In January 2008 actual revenues fell short of projections by $43 million. This came after the downward revision of revenue projections in December, and is a source of concern that revenues are being affected by the slowing economy.