Corporate Income Tax Forecast Methodology The corporate income tax forecast is Figure 6 CORPORATE INCOME TAX MODELS produced by: National Corporate 1. Forecasting total annual corporate ECONOMIC FORECASTS Profits Oregon Personal tax liability. Income 2. Forecasting annual liability by U.S. Personal Income TOTAL TAX payment type - advance payments, Historical Liability and Collection Data LIABILITY MODEL final payments, delinquent payments3 and refunds. FORECASTER JUDGEMENT 3. Convert the annual tax liability forecast by payment types into a ANNUAL LIABILITY BY quarterly collections forecast. PAYMENT TYPES Figure 6 outlines the different models and variables used to produce the QUARTERLY COLLECTIONS MODEL BY PAYMENT TYPE corporate income tax forecast. FORECASTER JUDGEMENT QUARTERLY COLLECTIONS FORECAST Corporate Liability Model The corporate income tax model is similar in nature to the personal models. However, the transition from collections to liability is far more complex. A specific corporate tax year may start any time during the same calendar year. Many corporations use calendar year or fiscal year as their tax year, but not all. As a result, collections and refunds for a given tax year are spread over several years (See Figure 7). The differing tax years also means Figure 7 that payments are received on Timing of Corporate Income Tax Payments multiple tax years simultaneously. Tax Year 1997 120 For example, an average of 72 Refunds Advance Final 100 percent of advanced payments 80 received during a calendar year 60 belongs to that tax year. The Millions of Dollars 40 remaining 28 percent belong to the 20 - prior tax year. (20) (40) The top half of Figure 8 shows (60) how collections data are processed (80) to develop monthly liability data (100) 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 by payment type. In essence, Months Since Start of Tax Year 1997 3 Delinquent payments are defined as advance payments associated with tax years that are at least two years prior to the current tax year and final payments for tax years that are at least three years prior to the current tax year. collections are split by tax year and then shifted back in time to create a liability data set. This produces a monthly liability variable for each payment type. The monthly data are then converted to annual data prior to forecasting each payment type. The main driver behind the total liability model is corporate profits. An Oregon specific forecast is not available. Instead, the ratio of Oregon personal income to U.S. personal income is used to derive Oregon’s share of national corporate profits. This process picks up overall growth in corporate profits as well as Oregon’s increasing share of profits. The total liability model also includes variables for recent Figure 8 corporate kicker refunds and the Collections 1991 recession. Split by Tax Year & Payment Type The total liability forecast is the main predictor of advance payments and refunds. Delinquent Advance Final Delinquent Refund payments tend to grow along with overall corporate profits. Final current year 1 year lag 1 year lag 2 year lag 2 year lag 3 year lag 1 3/4 year lag payments are calculated as total liability less advance and Shift collections data to line up by tax year delinquent payments plus refunds. and sum to get annual liability The individual equations are contained in Appendix B. Total Advance Final Delinquent Refund Liability = Liability + Liability + Liability - Liability Collections Model Corporate Liability Models The spreading equations take the annual liability forecasts by Collections Models payment type and convert them into monthly collections forecasts. Split liability by when collections occur The bottom half of Figure 8 and lag accordingly outlines this process. The steps involved in this process are laid out below: Advance Final Delinquent Refund 1. Annual forecasted liability by payment type is spread equally 72% current 28% 1 year 67% 1 year 33% 2 years 22% 2 years 78% 3 years 21 months over the months of the year. year in in in in in in collections future future future future future future 2. The liability series for all payment types except refunds is split into two pieces based on Seasonally adjust monthly collections data when collections for a given tax year’s liability occur. (The Aggregate monthly collections to quarterly collections reverse of the process used to convert collections to liability). This produces seven data series (see Figure 8). 3. The forecasts are then shifted forward in time to account for the time between when the liability is incurred and collections occur. 4. Historical seasonally patterns are applied to the forecasts using X-11 determined seasonal factors. 5. Monthly forecasts are summed to get quarterly collections forecasts. Forecaster Judgement The raw collections forecasts must be adjusted to account for tax law changes that are not included in the liability models. This includes recent legislation and policy actions such as adding more tax auditors at the Department of Revenue. Kicker credits that have yet to be taken are also reflected in this manner. In addition, recent collections trends must be taken into consideration. If there is reason to believe that collections patterns will vary from historical patterns, then the forecast can be adjusted accordingly outside of the forecasting models. Both of these items can have significant impacts on the final revenue forecast.