Monday, June 15, 2015

A graphic comparison of the Administration and ISER models ...


Over the weekend I worked on and published on "page one" of this blog a comparison and critique of two economic models designed to look at Alaska's fiscal situation ("The (much) better economic model for #AKFuture …").  

The first model was that released last weekend as part of Governor Walker’s “Conversations with Alaskans”. That model is available for download here (*see additional footnote at the end of the piece here).  The second was prepared by long-time Alaska economist Scott Goldsmith and released earlier this year by the University of Alaska-Anchorage’s Institute for Social and Economic Research, the best non-partisan economic think tank in the state. That model is explained and available for download here.

As I discussed at length in the weekend piece, there are significant differences between the two models and what they imply about Alaska's fiscal future.

After publishing the piece, yesterday a reader wrote that while they understood the points it would be helpful if I could find a way to demonstrate them graphically.  After thinking about it this morning I have worked the models and copied the results at the start of this piece.

The two graphs reflect the same base economic case to the extent the models permit.  The case incorporates: an FY 2016 oil price of $70 escalated thereafter by inflation; spending levels that start at an FY 2016 level of $5.3 billion, then decline to $4.83 billion in FY 2019 before starting to climb again at the rate of inflation plus 1% for population growth (which largely incorporates the analysis contained in Scott Goldsmith's most recent paper on sustainable budgets, "The Path to a Fiscal Solution: Use Earnings from All Our Assets" at p. 5); a constant inflation rate of 2.25 (the Administration's) and 2.3 (ISER) percent; contribution to state revenues of the portion of the permanent fund earnings reserve remaining after payment of the PFD and inflation proofing beginning in FY 2017 and various other factors as nearly equal between the two models as the options permit.

As the graphs show, despite the same (or nearly same) inputs, the results are startlingly different.  As I noted in this weekend's piece, the lead graphic in the Administration's model (the left one above) continues to show a chronic and substantial fiscal gap that users of that model are then encouraged to close through the use of additional, mostly tax-related but in some instances PFD limiting "revenue options."  

The ISER model, however, shows that the long term effect of simply adding the earnings reserve to the revenue stream is sufficient to close the gap, without the need for any more options.

As explained somewhat in yesterday's piece and in much greater detail in the series of ISER papers indexed here, the difference largely is in the scope of the vision incorporated into the models.

In addition to providing users with the option to change the first year price of oil, the ISER model also permits users the additional option of looking forward and adding (or subtracting) levels or sources of oil (or gas) production which may occur in the future. Those additions are one of the two best sources of new revenue to state government (in the sense they add, rather than take dollars from the private Alaska economy).  In this morning's run, I mostly used the default setting for those options contained in the ISER model, changing only the long term production decline rate to 2%, which many have estimated is likely to be the result of the passage of SB 21.

The Administration’s model, on the other hand, largely fixes the forward looking oil production curve to reflect the Administration’s own, admittedly highly conservative view of future production levels.  As explained in the weekend piece, it also appears artificially to limit the ongoing contribution levels available to the general fund from the permanent fund earnings reserve.

As the results show, the effects of excluding these factors are substantial.

These limitations make the future deficit levels produced by the Administration's model appear much larger than they actually are likely to be, with the consequence (whether intentional or not) of making the user believe that they need to toggle more and increasingly heftier “revenue [i.e., tax] options” in order to close the future gap.  The ISER model, on the other hand, enables users to make much more realistic judgments about the need for new "revenue options."

Readers who have not previously caught up with this weekend's piece are encouraged to do so by clicking here.  Hopefully these graphics will help additionally make the same points.