Abstract
This chapter addresses the question of whether states can improve the accuracy of revenue forecasts by using more advanced time series and Bayesian vector autoregression (BVAR) forecasting methods. Using state revenue data from Virginia, we first estimate baseline forecasts using autoregression (AR) and vector autoregression (VAR). We present the theoretical case for estimating a BVAR model, and then we present and compare forecasts based on the AR, VAR, and BVAR. Although there are gains in forecast accuracy in using the BVAR, the BVAR is not a panacea for forecasting the extremely volatile corporate income tax revenue series.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Virginia is one of Governing Magazine’s “Best managed” states, winning awards from professional organizations for the high quality of its financial reporting and maintaining AAA bond rating for two decades (Conant 2010).
- 2.
- 3.
Although there are forecasts from the government of Virginia, we do not analyze and compare them with our model-based forecasts as the data we use are based on different definitions of the various revenue streams.
- 4.
- 5.
Krol uses real defense spending, real crude oil prices, real U.S. personal income, consumer price index, real California housing prices, and real California personal income. We use total vehicle sales, production of defense and space equipment, retail gasoline price, Virginia housing price index, Virginia personal income, and Virginia unemployment rate.
- 6.
The data are available from a historical perspective rather than available in real time. This distinction can be extremely important empirically and can alter the rankings of forecast models.
- 7.
- 8.
Differencing is one of several ways to obtain stationarity. Notably, a cointegrating combination obtains, and the properties of levels forecasts can be improved by taking cointegration into account; see Engle and Yoo (1987). However, further research has shown mixed success when taking cointegration into account. Christoffersen and Diebold (1998) along with Hoffman and Rasche (1996) find mixed results while Shoesmith (1992) as well as Baghestani and McNown (1992) find improved forecasting accuracy. See also Elliott (2006) for further discussion.
- 9.
- 10.
Empirical results should be sensitive to the choice of model estimation period. Theoretically speaking, earlier years of the model estimation period are less relevant to explaining more recent revenues and hence reduces the accuracy of forecasts. With higher frequency data series, this assertion can be empirically verified. It is also the case that empirical results are sensitive to the choice of forecast period. Forecasts in distant future time periods are less accurate than the time periods closer to the end of the model estimation period.
- 11.
In fact, publicly available revenue forecasts are made on an annual basis.
- 12.
According to Elliott and Timmermann (2016), “although the BVAR forecasts generally perform well, no single approach appears to be dominant across variables and forecast horizons, illustrating the need to adopt different approaches to different situations and perhaps also consider forecast combinations.”
References
Baghestani, H., & McNown, R. (1992). Forecasting the federal budget with time-series models. Journal of Forecasting, 11(2), 127–139.
Baum, C. F. (2013). VAR, SVAR, and VECM models: Lecture slides for EC 823. Department of Economics, Boston College (Spring). Retrieved from http://fmwww.bc.edu/EC-C/S2013/823/EC823.S2013.nn10.slides.pdf.
Becketti, S. (2013). Introduction to time series analysis using Stata. College Station: Stata Press.
Boyd, D. J., & Dadayan, L. (2014). State tax revenue forecasting accuracy: Technical report. Retrieved from http://rockinst.org/wp-content/uploads/2018/02/2014-09-30-Revenue_Forecasting_Accuracy.pdf.
Brandt, P. T. (2013). Bayesian VAR time series analysis. Lecture, ICPSR Summer Program Time Series Analysis: Advanced Topics Workshop, Ann Arbor, MI, July 29.
Brandt, P. T., & Williams, J. T. (2007). Multiple time series models. Thousand Oaks, CA: Sage Publications.
Carriero, A., Clark, T. E., & Marcellino, M. (2015). Bayesian VARs: Specification choices and forecast accuracy. Journal of Applied Econometrics, 30(1), 46–73.
Carriero, A., Mumtaz, H., & Theophilopoulou, A. (2012). Forecasting fiscal variables using Bayesian VARs with constant and drifting coefficients. Unpublished (PDF).
Christoffersen, P. F., & Diebold, F. X. (1998). Cointegration and long-horizon forecasting. Journal of Business & Economic Statistics, 16(4), 450–456.
Conant, J. K. (2010). Budget deficits in the states: Virginia. Public Budgeting & Finance, 30(1), 33–57.
Conant, J. K., Bunch, B., Duncombe, W., Lauth, T. P., Robbins, M., Simonsen, B., et al. (2012). State budgeting in the aftermath of the great recession: A comparative perspective. Municipal Finance Journal, 33(2), 1–33.
Elliott, G. (2006). Forecasting with trending data. In G. Elliott, W. J. Granger, & A. Timmermann (Eds.), Handbook of economic forecasting (Vol. 1, pp. 555–604). Amsterdam: North Holland.
Elliott, G., & Timmermann, A. (2016). Economic forecasting. Princeton, NJ: Princeton University Press.
Engle, R. F., & Yoo, B. S. (1987). Forecasting and testing in co-integrated systems. Journal of Econometrics, 35(1), 143–159.
Haas, M. P., & Knittel, M. J. (1997). Revenue forecasting in Michigan. In Proceedings of the Annual Conference on Taxation and Minutes of the Annual Meeting of the National Tax Association (pp. 313–318). Washington, DC: National Tax Association.
Hendry, D. F., & Hubrich, K. (2011). Combining disaggregate forecasts or combining disaggregate information to forecast an aggregate. Journal of Business & Economic Statistics, 29(2), 216–227.
Hoffman, D. L., & Rasche, R. H. (1996). Assessing forecast performance in a cointegrated system. Journal of Applied Econometrics, 11(5), 495–517.
Johnston, J., & DiNardo, J. (1997). Econometric methods (4th ed.). New York: McGraw-Hill.
Karlsson, S. (2013). Forecasting with Bayesian vector autoregression. In G. Elliott & A. Timmermann (Eds.), Handbook of economic forecasting (Vol. 2B, 1st ed., pp. 791–897). Amsterdam: Elsevier North-Holland.
Kilian, L., & Lütkepohl, H. (2017). Structural vector autoregressive analysis. New York: Cambridge University Press.
Krol, R. (2010). Forecasting state tax revenue: A Bayesian vector autoregression approach. Northridge: California State University. Retrieved from http://www.csun.edu/~hcecn001/published/BVAR_Forecast.pdf.
Lütkepohl, H. (2007). New introduction to multiple time series analysis. Berlin: Springer.
McNichol, E. C. (2014). Improving state revenue forecasting: Best practices for a more trusted and reliable revenue estimate. Center on Budget and Policy Priorities-CBPP, Washington, DC. Retrieved from https://www.cbpp.org/sites/default/files/atoms/files/8-7-14sfp.pdf.
McShea, M. F. (2011). Forecasting Virginia state revenues: A vector autoregression approach. Unpublished (PDF).
Parker, J. A. (2014). Vector autoregression and vector error-correction models. Unpublished manuscript, Department of Economics, Reed College (Spring). Retrieved from http://academic.reed.edu/economics/parker/s14/312/tschapters/S13_Ch_5.pdf.
Pew Charitable Trusts. (2015). Managing volatile tax collections in state revenue forecasts. Washington, DC. Retrieved from http://www.pewtrusts.org/en/research-and-analysis/reports/2015/03/managing-volatile-tax-collections-in-state-revenue-forecasts.
Robertson, J. C., & Tallman, E. W. (1999). Vector autoregressions: Forecasting and reality. Federal Reserve Bank of Atlanta Economic Review, 84(1), 4.
Shoesmith, G. L. (1992). Co-integration, error correction and improved medium-term regional VAR forecasting. Journal of Forecasting, 11(2), 91–109.
Sims, C. A., & Zha, T. (1998). Bayesian methods for dynamic multivariate models. International Economic Review, 39(4), 949–968.
Stata Corporation. (2013). Vector error correction models: Stata Corporation. Retrieved from http://www.stata.com/manuals13/tsvec.pdf.
Stock, J. H., & Watson, M. W. (2001). Vector autoregressions. Journal of Economic Perspectives, 15(4), 101–115.
Todd, R. M. (1984). Improving economic forecasting with Bayesian vector autoregression. Federal Reserve Bank of Minneapolis Quarterly Review, Fall.
Virginia Department of Planning and Budget. (2010). Frequently asked questions. Retrieved from http://dpb.virginia.gov/budget/faq.cfm.
Virginia Department of Planning and Budget. (2014). Overview of Virginia’s economic and revenue forecasting. File attachment to e-mail correspondence from John Layman, Chief Economist for State of Virginia.
Virginia Senate Committee Staff. (2006). Understanding Virginia’s budget process. Richmond. Retrieved from http://sfc.virginia.gov/pdf/understandingvirginiasbudgetprocess.ppt.
Yaffee, R. A. (2010). Forecast evaluation with Stata. Paper presented at the United Kingdom Stata Users Group Conference, London School of Hygiene and Tropical Medicine.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 The Author(s)
About this chapter
Cite this chapter
McShea, M., Cordes, J. (2019). Forecasting Post-Crisis Virginia Tax Revenue. In: Williams, D., Calabrese, T. (eds) The Palgrave Handbook of Government Budget Forecasting. Palgrave Studies in Public Debt, Spending, and Revenue. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-030-18195-6_9
Download citation
DOI: https://doi.org/10.1007/978-3-030-18195-6_9
Published:
Publisher Name: Palgrave Macmillan, Cham
Print ISBN: 978-3-030-18194-9
Online ISBN: 978-3-030-18195-6
eBook Packages: Economics and FinanceEconomics and Finance (R0)