Government Budget Forecasting

Theory and Practice

Jinping Sun, Thomas D. Lynch

• Publisher: Routledge
• Year: 2017
• Print ISBN: 9781420045826
• eBook ISBN: 9781351565110
• Topic: Politics & International Relations
• Subtopic: Politics
• Index: Politics & International Relations

Revenue and expenditure forecasting plays an important role in public budgeting and financial management, particularly during times of financial constraint, when citizens impose greater accountability upon government to use taxpayer dollars more efficiently. Despite its significance, revenue and expenditure forecasting is often overlooked in the budget process, and there is an imbalance between practice and research in this area. Based on the collaboration of budget scholars and practitioners, Government Budget Forecasting fulfills two purposes:

1. Enhances the understanding of revenue and expenditure estimation both theoretically and practically
2. Stimulates dialogue and debate among practitioners and academicians to identify good forecast practices as well as areas for improvement

Divided into four parts, this comprehensive reference first examines forecast practices at the federal, state, and local levels, drawing on case studies that include California, Texas, and Louisiana. It then explores consensus systems and risk assessment, considering political factors and the costs of forecast errors. The text concludes with a call to transparency and guidance from a code of ethics, and a look at forecasting practices in emerging countries.

Chapter 1
Government Budget Forecasting: Improving the State of the Art

Jinping Sun and Thomas D. Lynch

Contents

Overview of the Book

Common Themes

Future Research

References

Revenue and expenditure forecasting are an integral part of government budget process and play an important role in public budgeting and financial management. Despite its significance, budget estimation is often overlooked in the literature. The focus of most public budgeting and finance books is primarily on budgetary politics, processes, or financial management, whereas revenue and expenditure forecasting are covered in certain chapters or sections in some texts (Axelrod 1995, Golembiewski and Rabin 1997, Lee et al. 2003, Lynch and Martin 1993, Mikesell 2003, Rabin 1992, Rabin et al. 1996, Steiss and Nwagwu 2001), and are “reduced to a minuscule or non-existent topic” (Frank and McCollough 1992, p. 1683) in others (Cozzetto et al. 1995, Rubin 2006, Thompson and Green 1998, Wildavsky and Caiden 2003).

Since the 1990s, the situation has improved to a certain extent, with journal articles covering different aspects of revenue or expenditure forecasting. Yet, there is a lack of comprehensive, systematic texts on the theories and practices of budget estimation in the public sector. This book intends to fill the gap by presenting the state of the art of government revenue and expenditure forecasting based on the collaboration between scholars and practitioners.

Specifically, this book has two purposes. The first is to help those interested in public budgeting and finance understand how revenue and expenditure estimation are done theoretically and practically. The second is to stimulate the dialogue and debate among practitioners and academics, so that good forecast practices can be identified and recommendations can be made to enhance revenue and expenditure estimation.

Overview of the Book

This book is divided into four parts. Part 1 covers the theories and practices of revenue and expenditure forecasting at different levels of government in the United States.

■ Rudolph Penner describes the federal revenue forecasting process used by the Congressional Budget Office and speculates about the reason for its forecast errors that are serially correlated. He concludes that the uncertainty inherent in the forecasts plays a major role in budget policy debates.

■ At the state level, Katherine Willoughby and Hai Guo, using data complied from the 2004 Government Performance Project survey, present an overview of revenue forecasting in U.S. state governments. They find that states using multiple methods including simple trend analysis, linear regression, or consensus forecasting tend to achieve greater accuracy in forecasting revenue, particularly for individual and corporate income taxes and general sales taxes.

Concerning individual states, James Richardson traces the evolution of revenue forecasting in Louisiana, as a part of its political history, from the boom and bust of the oil and gas industry in the 1970s and 1980s, traditional method of funding state government in the 1990s, to the aftermath of two major hurricanes (Katrina and Rita) in 2005. In addition, the role of the Revenue Estimating Conference in the state’s budget process is emphasized. Jon David Vasché et al. provide a comprehensive account of how state revenue and expenditure forecasting are done in the State of California. They examine major entities involved in the forecasting process and specific modeling used in developing estimates of different revenue sources (such as personal income tax, sales and use tax, and corporation tax) and spending areas (such as K-14 education).

Regarding the estimation of individual revenue sources, Xu et al. and Stinson et al. explain in detail how to forecast personal income tax in New York and Minnesota, respectively, by breaking down the components of income tax forecasts in each state and examining in depth the models used in each step. In addition to personal income tax, casino gaming has been legalized in 20 states to augment state revenues. Focusing on the State of Indiana, Jim Landers provides an overview of casino gaming activity in Indiana and discusses specific methods and issues in forecasting casino tax revenue.

On the expenditure aspect, Shiferaw Gurmu and William Smith discuss various approaches to forecasting welfare caseloads with an emphasis on Temporary Assistance for Needy Families (TANF) program, and apply them to Georgia TANF data. They also conduct short- and long-term forecasts for TANF caseloads using a dynamic model under different specification choices, and assess the accuracy of the projections.

To evaluate how states perform in budget estimation, Jinping Sun studies the revenue forecasting process and performance of three major forecasting agencies in New York. She concludes that the state’s revenue forecasting process meets the majority of criteria established by national professional organizations; the three major forecasting agencies did a good job of accurately forecasting state revenues from fiscal year (FY) 1995–1996 to FY2002–2003; and the three agencies’ forecasts are good by other criteria such as credibility, timeliness, and helping improve decision making.

■ At the local level, John Wong develops a methodology for small and medium-sized communities to estimate the base of a new local sales tax using detailed census data. This method is applied to the City of Derby, Kansas, and is 97.7 percent accurate in forecasting taxable retail sales.

Craig Kammholz and Craig Maher conduct a case study of revenue forecasting in the City of Milwaukee, Wisconsin, where the official revenue forecasting responsibility rests with the Comptroller’s office. They find that this arrangement not only provides additional resources for revenue estimation, but also protects the city from political manipulation. Further, Milwaukee performs well when compared to nine peer cities in terms of forecast accuracy and bond ratings.

To evaluate local government revenue forecasting, Christopher Reddick surveys city government finance directors in Texas. The results indicate that revenue forecasting is mainly an internal process and there is little participation from citizens or city council. In addition, cities have a small forecasting staff and typically use few prior years of data and expert and trend forecasting for revenue estimation.

■ With a different focus, Aman Khan introduces a comprehensive model to forecast the financial condition of a government’s enterprise operation based on its assets, liabilities, and net assets (fund balance) situation. Daniel Williams looks into data preparation for forecasting with the belief that well-prepared data can help in getting reliable forecasts. Basic steps such as data editing, adjusting for inflation, and dealing with seasonality are discussed in detail in this chapter.

Consensus budget forecasting is commonly used across the states, and chapters in Part 2 pertain to this practice.

■ William Earle Klay and Joseph Vonasek attempt to explore why consensus forecasting contributes to greater accuracy. They examine theories including the questioning of underlying assumptions and combining of forecasts, and present a historical study of consensus forecasting in the State of Florida—a state with more than three decades of experience in consensus revenue and expenditure forecasting.

■ Yuhua Qiao, on the other hand, conducts a telephone and e-mail survey of budget offices in 27 states regarding the extent of use, implementation, and performance of consensus revenue forecasting. She finds that states vary in how to implement consensus revenue forecasting in terms of structure, legal basis, funds to cover, and binding abilities. Consistent with Klay and Vonasek’s findings, consensus forecasting can improve forecast accuracy.

■ The significance of consensus forecasting is further augmented in two case studies. John Mikesell’s study of Indiana’s consensus revenue forecasting system reveals that rather than overemphasizing sophisticated forecasting methods, a politically balanced, transparent, and trusted process can produce accurate revenue forecasts that all participants, regardless of their political affiliation, accept as the base for budget appropriations. John Wong and Carl Ekstrom present an overview of the consensus revenue estimating process in the State of Kansas, including institutional arrangements for estimating state government revenues, and specific techniques used by individual Consensus Revenue Estimating Group members in economic and revenue forecasts. They conclude that the consensus process brings professionalism and more rigorous analysis to revenue forecasting and improves forecast accuracy.

There are always uncertainty and risks involved in budget estimation. Part 3 illustrates how to reduce uncertainty and mitigate risks in budget forecasting.

■ Xu et al. present methods for assessing forecast risks (including Monte Carlo simulation and fan charts), introduce symmetric and asymmetric forms for the forecaster’s loss function, and discuss how to choose an optimal forecast under a given loss function and distribution of risks.

■ Fred Thompson and Bruce Gates discuss risk management tools that states can use to achieve structural fiscal balance and manage cyclical fiscal imbalance, which include Monte Carlo simulation, present value cash flow analysis, target budgeting, portfolio analysis, hedging, self-insurance, and self-insurance pools based on simple mean-variance analysis. They argue that these tools are better than shifting financial obligations to other jurisdictions, borrowing from enterprise and trust funds, and other approaches employed by state governments. They then use the case of Oregon to demonstrate how these tools can help governments manage financial risks.

■ To aid budget forecasting and analysis, Ray Nelson proposes an integrated methodology that combines theories from tax policy and financial market risk management literature, and considers state sales, income, business, and other tax revenues as a portfolio. This methodology allows forecasters to better assess and predict alternative business cycle scenarios, and helps policy makers assess the implications of tax changes on base revenue levels and noncyclical and cyclical growth.

Other topics related to budget forecasting are discussed in Part 4.

■ Two chapters concern ethics of budget forecasting. Robert Smith identifies ethical dilemmas in budget forecasting in the public sector, examines their relationship to ethics principles, and presents a code of ethics as guidance for budget forecasting. Charles Garofalo and Nandhini Rangarajan explore the role of transparency in the ethical environment of revenue estimation. They examine individual and institutional resistance to increased transparency and propose three approaches to increase transparency in revenue forecasting: acknowledging the moral agency of revenue forecasters, creating criteria for deciding information disclosure, and adopting a consensus forecasting.

■ Different from forecast practices in the United States, budget estimation in other countries has its own contexts and characteristics. Sally Wallace develops a methodology to integrate fiscal architecture (impact of economic, demographic, and institutional changes) into budget forecasting and examines the impact of these changes on revenue forecasting and expenditure needs. She applies this methodology to India to show how it helps in improving the accuracy of budg...

Table of contents

• Cover
• Half Title
• Dedication
• Title Page
• Copyright Page
• Table Of Contents
• Acknowledgments
• Editors
• Contributors
•   1. Government Budget Forecasting: Improving the State of the Art
• PART 1: Forecast Practices
• PART 2: Consensus Forecasting
• PART 3: Uncertainty and Risk Assessment in Budget Forecasting
• PART 4: Other Topics in Budget Forecasting
• Index