• Print
  • Email
  • Share

Chicago Fed Letter, No. 365, 2016
The Illinois Budget Crisis in Context: A History of Poor Fiscal Performance

The author looks back at Illinois’s expenditure and revenue performance since the late 1980s to understand the sources of its current fiscal crisis. The article shows that compared with the national average, Illinois used to be a relatively low-expenditure, low-revenue state. This changed in the mid-1990s, when, unlike the typical U.S. state, Illinois began consistently spending more than it brought in. A major contributor to this budgetary imbalance was the accumulation of pension liabilities.

The state of Illinois is in the midst of a fiscal crisis. It went through the entire previous fiscal year (FY)1 without a budget and began the current fiscal year (in July) with a spending plan through only the end of the calendar year. Lawmakers cannot agree on how to solve Illinois’s financial problems, which are substantial: They include a $5.5 billion operating deficit, $7.7 billion in unpaid bills, and an estimated $111 billion in unfunded pension liabilities. The crisis has also led credit rating agencies to give Illinois the lowest ratings of any state—which has pushed up borrowing costs.2

How did Illinois get to this point? To address this question, I compare Illinois’s expenditure and revenue performance since the late 1980s with that of other U.S. states. I show that Illinois used to be a relatively low-expenditure, low-revenue state. However, unlike the spending of the typical U.S. state, starting in the mid-1990s, Illinois’s spending consistently outpaced its revenues (which have stayed low). In particular, the accumulation of pension liabilities has become a major source of Illinois’s fiscal problems.

There are a number of challenges in making a valid comparison of the fiscal situation in Illinois with that in other states. First, each state’s budget makes different distinctions between activities funded through its general fund and activities funded through special funds. Second, each state divides responsibilities for its activities differently between state and local governments. For example, some states fund K–12 schools primarily through state revenues, while others fund them primarily through local revenues. Third, states vary in population and income per capita. To address these challenges in comparing states, I use data from the U.S. Census Bureau’s annual Survey of State and Local Government Finances,3 which combines state and local government finances together and categorizes expenditures and revenues consistently across states. To account for differences in population and per capita income across states, I express expenditure and revenue figures as a percentage of gross state product (GSP), which is the most comprehensive estimate available of the overall size of a state’s economy.4

I first look at how Illinois compares with other states in terms of total state and local government expenditures (as defined by the Census Bureau). Panel A of figure 1 shows expenditures as a percentage of GSP for Illinois and the average of this measure for all U.S. states since FY1988. According to the Census Bureau data, Illinois was and continues to be a below-average spender, though the difference in its spending relative to that of the typical state (shown in gray) has been steadily shrinking since the early 1990s. The gap was as high as 3.7 percentage points in FY1989, but had fallen to 1.2 percentage points by FY2013.

1. State and local government expenditures as a share of GSP

 Figure 1

Notes: GSP means gross state product. Pension liability data for Illinois come from the Illinois Department of Insurance. These data are available over the entire period shown. Liabilities for all U.S. state and local governments come from the Board of Governors of the Federal Reserve System’s Financial Accounts of the United States. However, pension liability data are not available for individual states prior to fiscal year (FY) 2001 from this source, so the average change in pension liabilities as a percentage of GSP before FY2001 cannot be calculated. Instead, the change in total pension liabilities for all U.S. state and local governments as a percentage of U.S. gross domestic product is calculated for FY1988–2013.
Sources: Author’s calculations based on data from the U.S. Census Bureau (see note 3); U.S. Bureau of Economic Analysis from Haver Analytics; Board of Governors of the Federal Reserve System from Haver Analytics; and Illinois Department of Insurance.

One important aspect of state and local governments’ fiscal policy not fully tracked by the Census Bureau is their defined benefit pension systems. To understand what is missing from the Census Bureau data requires a brief explanation of how pension systems generally work. When government employees retire, for the rest of their lives they receive a guaranteed monthly check (based on the number of years worked) from the system. While these employees are still working, they and their government employer make yearly contributions to the system to cover the additional future benefits earned for an additional year of work. The required contribution is calculated based on actuarial assumptions about how long retirees will live and what the returns will be from investing the contributions.

I now consider two approaches to translating how a pension system operates into an income statement accounting framework for state and local governments. The first approach treats the pension system as separate from other government functions. In this case, the yearly change in a system’s liabilities is counted as expenditures and the yearly change in a system’s assets is counted as revenues. There are three factors that determine the change in a pension system’s liabilities: the amount of new benefits earned by employees for an additional year of work (which are always positive expenditures), changes in the benefits formula (positive or negative expenditures), and changes in actuarial assumptions (positive or negative expenditures). There are also three factors that determine the change in a pension system’s assets: the contributions of the government and employees (which are always positive revenues), returns from investing existing assets in the financial markets (positive or negative revenues), and payments to retirees (negative revenues).

The second approach integrates a pension system’s income statement with the rest of the state’s income statement, which is what the Census Bureau does. This requires a few changes: First, while the government’s contribution is revenue for the pension, it is also a governmental expenditure, so it is excluded from the tally of both total expenditures and revenues. Second, payments to retirees are counted as expenditures rather than negative revenues. Finally, changes in pension liabilities are not tracked (this changed recently), so they are missing from the Census Bureau’s tally of total expenditures. Because pension liabilities have been growing over time, this means that the Census Bureau data understate state and local governments’ true expenditures.5

Panel B of figure 1 shows expenditures as a percentage of GSP amended to include changes in state and local government pension liabilities. Both the U.S. state average and Illinois lines are higher in panel B than in panel A because pension liabilities have been consistently growing (the average annual increase over the period FY1988–2013 was 1.6% of gross domestic product for the U.S. states and 2.0% of GSP for Illinois6). As in panel A of figure 1, the difference between the average U.S. state and Illinois has steadily declined since the late 1980s, though now the difference narrows faster. In recent years, the difference has approached zero, indicating that it is no longer appropriate to consider Illinois a low-expenditure state.

As a low-expenditure state (prior to more recent times), Illinois could also afford to be a relatively low-revenue state and still maintain a balanced budget. Figure 2 shows that Illinois has consistently had lower revenues as a percentage of GSP than the average U.S. state.7 However, in comparing figures 1 and 2, it becomes clear that while the gap in expenditures between the average U.S. state and Illinois has shrunk to close to zero, the gap in revenues has only shrunk a little. Figure 3 makes the comparison explicit and shows that starting in the mid-1990s, Illinois began spending a greater percentage of its revenues than the average U.S. state. In addition, since the late 1980s, Illinois has been spending more than 100% of its revenues, meaning it hasn’t had a truly balanced budget for well over two decades and has been accumulating debt.

2. State and local government revenues as a share of GSP

 Figure 2

Notes: GSP means gross state product. U.S. Census Bureau data include pension revenues by individual state for fiscal years 1988–2013.
Sources: Author’s calculations based on data from the U.S. Census Bureau (see note 3) and U.S. Bureau of Economic Analysis from Haver Analytics.

3. State and local government expenditures as a share of revenues

 Figure 3

Notes: Expenditures include the change in pension liabilities. Pension liability data for Illinois come from the Illinois Department of Insurance. These data are available over the entire period shown. Liabilities for all U.S. state and local governments come from the Board of Governors of the Federal Reserve System’s Financial Accounts of the United States. However, pension liability data are not available for individual states prior to fiscal year (FY) 2001 from this source, so the average change in pension liabilities as a percentage of revenues before FY2001 cannot be calculated. Instead, the change in total pension liabilities for all U.S. state and local governments as a percentage of total revenues for U.S. state and local governments is calculated for FY1988–2013.
Sources: Author’s calculations based on data from the U.S. Census Bureau (see note 3); Board of Governors of the Federal Reserve System from Haver Analytics; and Illinois Department of Insurance.

How, specifically, has Illinois overspent relative to its revenues? To shed light on this question, I look at expenditures by category, as classified by the Census Bureau, for Illinois versus the average U.S. state during FY1994–2010 (Illinois’s overspending became markedly higher than that of the typical U.S. state in FY1994, and Illinois raised taxes in FY2011). For those 17 years, I calculate the average of yearly expenditures as a percentage of total revenues. Figure 4 shows that over those years, Illinois’s spending averaged 115.9% of its revenues compared with 105.7% for the typical U.S. state. Thus, while spending outstripped revenues for both Illinois and the typical state during this span, Illinois overspent by much more (10.2 percentage points). The top two categories on which Illinois spent more were the change in pension liabilities (4.6 percentage points more) and employee retirement (2.8 percentage points more). Pension-related spending, then, makes up almost three-quarters of the difference between Illinois’s spending and that of the typical state.8 Illinois also spent 1.3 percentage points more of its revenues on general debt interest than the average state—a sign that Illinois was accumulating debt outside of its pension system as well.

4. Average annual expenditures as a share of total revenues, by category, FY1994–2010

 Figure 4

Notes: All data except pension liability data come from the U.S. Census Bureau and are available for individual states for fiscal years (FY) 1994–2010. Pension liability data for Illinois come from the Illinois Department of Insurance. These data are available for FY1994–2010. Liabilities for all U.S. state and local governments come from the Board of Governors of the Federal Reserve System’s Financial Accounts of the United States. However, pension liability data are not available for individual states prior to FY2001 from this source, so the average change in pension liabilities as a percentage of revenues before FY2001 cannot be calculated. Instead, the change in total pension liabilities for all U.S. state and local governments as a percentage of total revenues for state and local governments is calculated for FY1994–2010. The values reported in the figure may be subject to rounding error.
Sources: Author’s calculations based on data from the U.S. Census Bureau (see note 3); Board of Governors of the Federal Reserve System from Haver Analytics; and Illinois Department of Insurance.

Conclusion

This article reveals that Illinois’s fiscal crisis has been a long time coming. From the late 1980s on, Illinois has spent more than it has collected in revenues. And while the typical U.S. state has also generally spent more than it has collected, Illinois’s overspending has outpaced the national average since the mid-1990s, primarily through pension spending. How could Illinois get away with this for so long when it is required by law to have a balanced budget? Over the years, lawmakers used a variety of techniques to put off paying the bills, including underpaying into the pension systems. Such techniques can work for only so long, and Illinois is now coming to terms with over 20 years of poor fiscal performance.


1 The fiscal year for Illinois ends in June and is denoted by the later year. For example, fiscal year 2016 started in July 2015 and ended in June 2016.

2 See Elizabeth Campbell, 2016, “Illinois awash in red ink as budget deal buys it six more months,” Bloomberg, July 11, available online by subscription.

3 Available online. I accessed the data via the Urban Institute and Brookings Institution’s State and Local Finance Initiative Data Query System (SLF-DQS).

4 Other common approaches to adjusting for differences in population and income across states include presenting the expenditure and revenue figures in per capita terms or as a percentage of state total personal income.

5 One wrinkle in the case of Illinois’s pension system is that the state sold pension obligation bonds in FY2003, FY2010, and FY2011 (some of the proceeds from the sale in FY2003 were not put into the pension system). While the Census Bureau treats proceeds from pension obligation bonds as revenues when analyzing pension systems separately from other government functions, it does not treat proceeds from bond sales for any type of bond (including pension obligation bonds) as revenues in the integrated accounting framework. Thus, Illinois’s use of pension obligation bonds does not require any additional adjustment to the data presented in this article. Note, however, that bond proceeds will affect later changes in the market value of pension assets and that those later changes will be offset by interest payments on the pension obligation bonds, which are included in the interest on general debt expenditure category (see figure 4).

6 Author’s calculations based on data from the U.S. Bureau of Economic Analysis from Haver Analytics; Board of Governors of the Federal Reserve System from Haver Analytics; and Illinois Department of Insurance. The spike in Illinois expenditures in FY1998 (as seen in panel B of figure 1) is the result of increases in benefits for beneficiaries.

7 The sharp declines in revenues in FY2002 and FY2009 (as seen in figure 2) were the result of nationwide recessions.

8 Note that pension spending is a form of compensation for government employees, so it is appropriate to think of this spending as spread out among the other expenditure categories based on their employment shares.



Opinions expressed in this article are those of the author(s) and do not necessarily reflect the views of the Federal Reserve Bank of Chicago or the Federal Reserve System.



Federal Reserve Bank of Chicago, 230 South LaSalle Street, Chicago, Illinois 60604-1413, USA. Tel. (312) 322-5322

Copyright © 2018. All rights reserved.

Please review our Privacy Policy | Legal Notices