Tax revenues or PILOT II

Posted by Brent Sohngen

Please feel free to email with any questions: sohngen.1@osu.edu

The upshot of this post (to see the last post referred to below, go here)

• You can’t compare two financial analyses that use fundamentally different input data. Buckeye Institute and I used different data.  I got mine directly from Open Road Renewables, and it was the input data they used for their Ohio Public Siting Board reporting. Buckeye Institute claims to use the same data, but it’s not. You can see the differences below.
• The interest rate discussion is fascinating, but ultimately is a decision folks in Knox County have to make. I provided the results at multiple interest rates in my previous blog post. Here, I review the rates Knox County has obtained in recent years for money they have invested ($40 million in total) and money they have borrowed. It’s a simple review of their actual past decisions that tells you what their actual discount rates have been and currently are.  Hint: They invest most of their money in CDs, at what looks to be around 1.8% – 2.0%, although some recent investments in other vehicles may be as high as 5.5%. They borrow money at 2.0% to 4.0%.
• As an academic economist, I don’t have a stake in what Knox County chooses to do with respect to the PILOT versus tax revenue approach. It’s a complicated decision, made more complicated by confusing data presented by multiple parties. My hope, though, is that since the discussion is mainly about data, I have been able to clarify why different results have emerged.

Data Issues

Open Road provided me a set of projected PILOT payments and tax revenues over a 40-year period, which they used in their OPSB analysis. Buckeye Institute made a good point that you need to discount them and account for inflation before comparing. They then did this discounting. I decided to look at the numbers, so I asked Open Road to send me the exact numbers they used in their OPSB filing, which I then discounted myself.  Those results – discounted and undiscounted numbers – were in my previous blog post.

In theory, one can compare what I did with what Buckeye Institute did, but it turns out that’s not so easy because of the way Buckeye Institute presents their data and the impossibility of getting them to provide their inputs. Buckeye Institute claims they got the data from Open Road’s filing, however, they don’t provide the original undiscounted data on their website. What they do provide is the discounted data. I requested the original data from them multiple times, and they refused to send it. I could only email with an administrative person there, who was very nice, but none of the people who write blog posts would bother responding.

However, since I can do math, I worked backwards from their discounted data to the original data using their assumptions.

The problem is when I worked backwards with the Buckeye Institute data, I found that their basic numbers differed from the ones used by Open Road in their filing. Below are figures of the two sets of data graphically. At the heart of this issue, Buckeye Institute does not appear to be using the same initial data used in the Open Road OPSB filing. I have no idea why this difference exists. Since Buckeye Institute won’t even respond to my simple data request, I don’t know that I’ll ever find out.

I would recommend that people look carefully at this difference before concluding anything from the analysis done by Buckeye Institute. Buckeye Institute is using different data, so is conducting non-comparable analysis, which may or may not reflect the actual differences between the PILOT payments and the property tax revenues.

 

Discount rates

Then there is the discount rate. Again, from the economist’s perspective, it’s not in my purview to say what discount rate Knox County should use. Thus, I presented my results over several assumptions.

It is possible, though, to say something about what discount rates Knox County actually does use. For this, I went to their annual financial report for 2022, which I found online: https://www.knoxcountyauditor.org/media/12670/knox-county-annual-comprehensive-financial-report-2022.pdf.

It turns out they use lots of different interest rates, depending on the situation. They have annual revenues of around $68 million. They don’t spend all their revenue at once, and have a portfolio of savings, which are invested in a range of assets (see note 6 in their report), amounting to about $40 million. They don’t report the interest rates on the investments, at least as far as I could find, but I looked at many of the investments online to see what kinds of rates are likely.

Their biggest investment, nearly half of it, is in negotiable CDs. The national average rate for CDs was 0.2% per year for a 1 month CD up to 1.86% per year for a 12 month CD (see https://www.fdic.gov/resources/bankers/national-rates/index.htm). I know you can find higher rates online for a CD at your favorite bank.  Those are introductory used to draw you in. Knox county lists its CD rates as negotiable, so they may be getting more than the national average. They can provide this information.

The second biggest investment they have, nearly $7 million, is something called FHLB, which stands for Federal Home Loan Banks. They recently issued bonds in the 3-6% range. The maturity of the bonds Knox county has invested in are 1 to 5 years, so many of the bonds that Knox county holds have much lower rates since they were invested more than a year ago.  Rates started increasing in late 2021. Before that, FHLB bond rates were in the 1% to 3% range.

The county also invests in STAR Ohio, which stands for State Treasury Asset Reserve. The rate of return on this asset is 5.5% in January 2024. A year ago, the rate of return was 4.5%, but two years ago it was 0.8%. In 2019, returns were around 2.5%. The last time rates were this high was in the early 2000s (see https://starohio.tos.ohio.gov/).

One can also look at the funds the county borrows. For this, I looked at long-term liabilities, which can be found in note 18. They are borrowing funds at rates of anywhere between 0% and 4% based on these data. The 0% rate applies to some loans from the state or federal government. But they do have some borrowing at 0%. Apparently, even counties have parents. Who knew?  A more realistic range is 2.0% to 4.0%.

Based on this analysis, it looks like Knox County invests money for returns of 1.8% to 5%, with the majority of their investments in safe, but lower yielding CDs returning 1.5% to 2%.

It looks like they borrow money at 2%- 4%.

What’s the right rate for analysis? It’s probably not 7%, but could be 2% to 5%. Did I mention my previous analysis? That range appears there (interestingly, if you look at the Knox County annual reports from 2007 to 2022, you’ll see a whole lot more of 1-2% interest rates than 5-7% interest rates).