The Alaska State Legislature is currently considering a bill that would allow the Department of Revenue (DOR) to borrow money to make oil tax credit purchases. The bill is called HB 331 in the House and SB 176 in the Senate. Both bills sit in their respective finance committees, but are said to be a part of the final compromise to get done with the session.

The entire issue is very touchy and ripe for political rhetoric. I’m not going to get into all that.

This quick analysis doesn’t speak to the politics of the issue, or the legal merits that have been raised. The only thing this analysis contemplates is the money.

I’ll give you the conclusion up front. If you are committed to purchasing these credits, this bill is a good deal for the State (from a pure cash flow basis). I feel like I need to explain that caveat, so I’ll come back to it at the end.

Tax Credit Purchases

Here’s why. According to a presentation DOR gave to the House Resources Committee last month, there are a little more than $1 billion worth of tax credits that are eligible for purchase.

Cash Out Option

Let’s assume the legislature wanted to purchase them all right now. In order to do that, they would need to draw down the Earnings Reserve Account (ERA) of the Permanent Fund by that billion dollars (this is an oversimplification, but it doesn’t change the outcome).

Statutory Draw

Instead, let’s say that they decide to purchase them over time, according to the statutory formula (here’s that caveat again, hold on).

In this scenario, they appropriate $184 million out of the ERA this year (they will probably say this money is coming from the general fund, but that just shifts another general fund item to be drawn off the ERA. So, we may as well call it what it is).

By only paying the formula amount, it leaves $847 million in the fund ($1,031 – $184). Those funds will earn interest while they sit in the account. In fact, they most likely earn somewhere between $50 and $100 million in interest given the Fund’s past performance.

Then, in next year’s budget, let’s assume they will again follow the formula. This means an appropriation of $168 million for FY20 purchases. Because of the money left in the bank, about half of those purchases will be made from the interest that was earned by not purchasing all the credits in the year prior.

If you follow this logic, you get something that looks like this:

		Statutory Formula
FY	Beginning Balance (7/1)	Interest Earned (Avg. of 50,000 simulations)	Appropriated for Next Year Credit Purchases
2019	$847	$87	$168
2020	$766	$79	$168
2021	$676	$69	$167
2022	$579	$59	$170
2023	$468	$48	$174
2024	$342	$35	$0
2025	$377	$39	$0
2026	$416	$43	$0
2027	$459	$47	$0
2028	$506	$52	$0
2029	$558	$57	$0
2030	$615	$63	$0
2031	$678

So, by 2031 you have $678 million more in the bank than if we just purchased all the credits today.

HB 331 / SB 176

But the real question is this: How does the status quo compare to the proposal being offered?

Well, I did that math too:

		Bonding
Fiscal Year	Beginning Balance (7/1)	Interest Earned (Avg. of 50,000 simulations)	Debt Service Payments Required for Next Year
2019	$1,004	$103	$30
2020	$1,077	$111	$66
2021	$1,123	$115	$96
2022	$1,141	$117	$118
2023	$1,141	$117	$128
2024	$1,130	$116	$128
2025	$1,119	$115	$128
2026	$1,106	$113	$128
2027	$1,092	$112	$128
2028	$1,076	$110	$72
2029	$1,115	$114	$37
2030	$1,192	$122	$13
2031	$1,301

By the end of the bond maturation period, there’s more money in the bank than we started with (and about $600 million more than the statutory formula scenario).

It’s kind of like clearing the credits for free, purchasing the credits with the extra interest we earned.

How Does This Work?

This happens because the borrowing rate for the State is something like 5% while the earnings in the ERA have averaged over 9% per year in the last 40 years. The difference between those interest rates is what bankers call “spread” and it is how they make money.

Risk

Of course, there is risk associated with this idea as well. Those debt service payments are locked in, but the ERA earnings are volatile. So, how big of a risk are we taking?

Well, when I run the simulation I get an average of 2.5 years during the 13 years under observation in which the Fund doesn’t make a positive spread. However, in the 10 years that it does, the spread is usually great enough to overcome those bad years.

When looking at the entire 13 year period, the State is worse off in less than 1% of the simulations.

The Upshot

If you’re on the fence about this bill based on cash flow and risk, jump to the support side. It makes sense to take advantage of our borrowing capacity to clear the deck immediately while keeping money in the bank to earn a spread. Chances are, doing so also leads to accelerated oil production, which has its own additional benefits.

If you have reservations about the legal construct of this bill, or if you question the entire idea of purchasing tax credits at all, I can’t change your mind.

Caveats

I keep saying “if we decide to make these purchases” because there is disagreement about the State’s obligation when it comes to these tax credit purchases. Some people think the statutory formula is being calculated incorrectly. Other’s point out that the entire program contains a “may” clause, and is therefore discretionary. Still others point to the recent Supreme Court decision which says that statutory formulas are subject to appropriation – so the legislature doesn’t have to pay anything at all unless it wants to.

Obviously, if you take one of these views, this whole analysis doesn’t matter.