More on the cost of the COLA adjustment (with correction)

By Steve Poftak
May 29th, 2008

CORRECTION: Jim Lamenzo found two errors, one major, in this post. First, a clarification, all this chatter refers only to state and teachers pensions, not locals. The first error — I said 2008 below, when I meant 2028.

Second, and this is a major one — I screwed up summing the figures from the PERAC 2005 report. Summing them correctly, gives you a range of $2.7b - $3.5b (not $4.7b - $5.5b). So, my original estimate was clearly too high. Its still substantial but below the original estimate. I regret the error.

I am, however, still sticking with my original point — that this is a large sum of money to be spending and should go through a transparent process.

ORIGINAL POST:

Buckle up folks, this is going to be a long, jargon-laden post.

I first came up with an estimate of the proposed COLA increase on this blog, taking the cheapest three years of the pension funding schedule ($3.8b) and the highest three years ($8b). Here’s my exact wording:

An additional three years of funding the state’s pension liability would require between $3.8 billion to $8 billion in additional operating budget expenditures.

I had a conversation with a reporter on the topic and noted that my calculation did not account for the time value of money, only the actual dollar cost over time, and was based on the information available to me at the time. The number subsequently was carried in the Globe.

I made a mistake in this estimate but I believe that the actual number remains within this range. PERAC, in the form of Jim Lamenzo, the state’s chief actuary, got in touch and pointed that my original calculation was including normal cost (in layperson’s terms — what the state should be paying in to cover the future costs of current employees; which is distinct from funds paid in to retire the unfunded liability) in my original calculations. Those should not be counted as part of the unfunded liability. He was kind enough to give me the figures for normal costs. With those taken out, my original estimate should have been between $2.7b - $5.5b.

I’m more confident now that the actual dollar cost (again over time and ignoring the time value of money) will be closer to $5b, well within the range of my original estimate. PERAC itself has a report from 2005 that calculates the incremental cost of funding the COLA increase under the current funding schedule and a second scenario funding it out to 2008. It is no longer 2005, but the numbers are directionally accurate and they put the cost in actual dollars at between $4.7b and $5.5b to fund the COLA increase.

Jim’s primary objection to my estimate was over the issue of the time value of money, which he believes overstates the case. He puts the present value of the additional liability from the COLA increase at roughly $1.1 billion. In other words, to fully fund the COLA increase, we would need to put $1.1 billion into the pension fund this year. He is correct, but…

My point is that we are not going to fund it immediately and we are going to fund it over time. I accept Jim’s point, and would use his number with a specific audience, but I’m trying to make a case to the broader policy community that the current outside sections will allocate roughly $5 billion in operating budget funds over the next 18 years that will not be spent on other items. I don’t think the process has had the transparency or public debate that is needed for a decision of this magnitude.

I regret the error in the initial estimate, but I’m more confident now that the correct estimate of operating budget expenditures (again, if you ignore the time value of money) is in the area of $5 billion and within the range of the original and corrected estimate.

For the few remaining readers, thanks for your patience and I’d appreciate any comments. Also, I’m going to give Jim a chance to respond if he’d like to.