Geoscience Reference
In-Depth Information
government had lost one-sixth of its ability to provide public goods and services—
education, safety, healthcare, and infrastructure—for Washington residents.
9
8.7
Open Economy
Tiebout was known for his quips. One dealt with social choice: “People vote with
their feet.” Another pertained to forecasting: “Forecasts are bound to be wrong, at
least to a degree.” The implication of the latter quip is that economists cannot
promise accurate predictions, only reasonable ones.
Nevertheless, a strong test of a regional economic model is the accuracy of its ex
ante forecasts. Since its inception, the prediction errors of the Puget Sound
Forecasting Model (Conway
2001
) have been tabulated and analyzed on a regular
basis. The most recent evaluation entails measuring one and 2-year-ahead errors for
the nineteen end-of-year forecasts made since 1993 (Conway
2013a
). The purpose
of this exercise is two-fold: to quantify the uncertainty of the predictions and to
improve the specification of the forecasting model.
10
The 100-variable econometric model generates forecasts for the Puget Sound
region, a four-county area comprising three-fifths of the Washington economy.
Variables are expressed in terms of their growth rates (the change in their natural
log) to enhance the statistical robustness of the model. Following the economic base
theory, employment in basic industries, such as durable goods manufacturing, is
predicted using exogenous variables (e.g., the U.S. industrial production index),
9
One concern about Washington's economic future is its dysfunctional state and local tax structure.
Counting retail sales taxes (which include taxes on new construction), business and occupation taxes
(much of which are passed on to consumers in the form of higher prices), and so-called sin taxes, the
state is twice as dependent on sales taxes as the rest of the nation. As a consequence, the tax system is
not only volatile but it is also regressive and inadequate. The Institute on Taxation and Economic
Policy (
2013
) rates Washington as having the most unfair tax structure in the nation. Inadequacy
refers to the inability of tax revenue to grow with the economy (as measured by personal income)
without expanding the tax bases or raising the tax rates. The problem of inadequacy is evident in the
forecasting equation for Washington taxable retail sales (Conway
2010
). With each variable
expressed in change-in-log form, the explanatory variables are personal income, the unemployment
rate, and housing permits. Trend growth in taxable retail sales is projected by personal income, while
cyclical changes are predicted by the unemployment rate and housing permits. While an adequate tax
base would have an income elasticity of 1.00, the estimated elasticity is only 0.80 (t-value
¼
14.7) in
the case of Washington taxable retail sales. This is the chief reason why Washington's state and local
effective tax rate (state and local tax revenue as a percent of personal income) fell from 11.6 % in FY
1995 to 9.7 % in FY 2011, putting it well below the 10.6 % 40-year average for all U.S. state and
local governments. This has created a tax policy bind: should Washington maintain the current tax
bases and tax rates, causing a further decline in the adequacy of state and local tax revenue; or should
it expand the tax bases and raise the tax rates, exacerbating the unfairness of its tax system?
10
One weakness of the Computable General Equilibrium (CGE) model, the most popular supply-side
model, is the difficulty of assessing its prediction errors. As Dixon and Rimmer (
2009
) put it, “how do
I know the results (of the CGE models) are accurate?” After assessing the forecasts of USAGE,
Dixon and Rimmer concluded that “USAGE output forecasts at the 500-commodity level comfort-
ably outperform trends. On the other hand, average errors for these forecasts seem alarmingly high.”
