3:30 PM Update:  The Governor’s Office has responded to Zycher’s observations below, and that response can be found here.

As we rush speedily into a vote tonight on a "Big-5" produces budget plan that includes nearly $15 billion in assorted tax increases, one of the major "concessions" that is in the plan, to "sweeten it" for GOP support is the placing of a Spending Limit Initiative on the ballot (though we note that the Limit, if it passes, also extends the tax increases).

One of the policy experts on economics and state finances that I look to in this area is Dr. Benjamin Zycher of the Pacific Research Institute.  Zycher was asked by the Howard Jarvis Taxpayers Association to review the final language.

The results were startling, and should be of concern to all Republicans- and I have to figure this is new information being interjected into the debate (it is unfortunate that things are being rushed — I am glad Dr. Zycher had the time and inclination to perform an analysis…

I should also add that this is the only analysis that we have on this spending limit.  Requests to get more information in writing on this budget deal, including the cap, from GOP leadership has been declined. 

Observations on the California Revenue Projection/Spending Limit Proposal

 Dr. Benjamin Zycher*
February 14, 2009

This note presents five summary observations on the revenue projection/implicit spending limit proposal soon to be considered by the California Legislature.  It discusses also, again in summary fashion, the general problem of imposing fiscal discipline in a way that creates incentives for interest groups and public officials to favor policies yielding economic growth stronger rather than weaker.  These observations are as follows:

• The degree to which the estimated revenue projection trend and the Budget Stabilization Fund would operate actually to limit general-fund spending is not clear.   Legislative staff members during informal discussions indicated that revenues above the official projected trend line (discussed below) are forced into the Budget Stabilization Fund, and otherwise may be used only for one-time items and the like.  But explicit language in the proposal allows the governor to suspend or reduce transfers into the BSF; and other explicit language allows the governor to declare an emergency and transfer funds into the general fund.  The constitutional definition of “emergency” may impose an effective limit on the latter gubernatorial power, but the former power to suspend or reduce transfers into the BSF does not appear to be constrained similarly.  Accordingly, the degree to which the revenue projection/BSF combination actually will reduce future spending growth is problematic. 

• The methodology for projecting revenues is highly problematic.  Because spending, notionally, is limited by revenues and by the required transfers into the BSF, the behavior of the 10-year projection methodology is crucial.  The language mandates a “linear regression of general fund revenues as a function of fiscal year” for the previous 10 fiscal years.  This is merely a trend line.  The use of only ten observations for such projection purposes is highly problematic; but more important, because of the mathematical algorithms underlying standard regression lines, the projections effectively will be driven not by ten observations, but instead by one or two observations at the beginning of the ten-year period, and by one or two observations at the end.  In short, the slope of projected revenue growth will be affected heavily by any outliers at the beginning and end of the ten-year period; outliers in the middle of the period might shift the whole trend line up or down, but that effect is less likely because the mid-period outlier would have to be very different from the larger revenue trend, an unlikely occurrence because revenues tend to be highly correlated over time.  Moreover, the likely use of nominal dollars (unadjusted for inflation) for projection purposes will yield serious problems if the inflation rate changes substantially after the projection period; in real terms the state could find itself with too much or too little real purchasing power.

• The language allows future tax increases not to affect the projected revenue trend, but does allow the assumed revenues from a tax increase to be included in revenues for the then-current fiscal year.  Legislative staff members have indicated that the provision excluding recent revenue legislation from the revenue projection is intended to allow for a “clean” revenue trend line.  But, therefore, it also allows for a future tax increase not to affect the revenue trend line adversely due to the negative economic effects of a tax increase.  But the ensuing language allows the purported revenues from a tax increase to be added to the projection for revenues for the (then-) current fiscal year.  Accordingly, the pursuit of a “clean” revenue projection—highly problematic in any event for the reasons noted above—may allow a future tax increase to yield (assumed) additional revenues and thus higher spending for the given fiscal year without affecting the following revenue projections because of the adverse effects of higher taxes on the tax base.  

• In particular, the proposed ten-year projection as a fiscal constraint will make future tax reduction very difficult.  The favorable future economic effects of tax reduction would be excluded from the projection methodology, but the (assumed) reduction in revenues would be recognized for the then-fiscal year. 

• Fiscal institutions should be designed to focus official attention on economic growth directly.  A spending limit is more useful if it is defined in terms of the size of the economy—a percentage of state gross product as an example—because such a limit yields disincentives to substitute regulation and other actions in place of general fund taxes and outlays.  A limit defined in terms of inflation and population would not have that effect, and for various reasons might have the opposite effect; and use of the ten-year revenue projection yields only the weakest of such incentives.  An alternative useful approach would be to limit outlays to some percentage (say, 105 to 110 percent) of revenues two years earlier.  Public officials and spending interests must have a stake in a California economy bigger rather than smaller.

This entry was posted on Saturday, February 14th, 2009 at 12:00 am and is filed under Blog Posts.