The HoCo By Design House of Cards

Introduction

The FY2022 Annual Comprehensive Financial Report (ACFR) states that at the end of the [FY2022] fiscal year, the County had total long-term debt outstanding of $2.1 billion of which $1.8 billion comprises debt backed by the full faith and credit of the government. The remainder of the County’s debt represents bonds secured solely by specific revenue sources (i.e., revenue bonds). The County’s total long-term debt increased by $158.5 million (8.2% during that year).

Even though the County issues new debt to finance projects related to new development every year and the debt has increased every year, the fiscal impact results suggest that residential development generates net revenue. This is clearly not the case since even the County assumes revenue from new residential development is used to pay off debt due to previous years' new residential development.

The following graphic summarizes a comparison of various projection approaches. The label of these approaches and their description are shown below. The result shows that if the revenues projected development are to cover their own cost, the County would face a shortfall of $450 to $500 million over 20 years.

County Funding Only: This case reproduces the County's results, by separating total school operating and capital expenditure into, State, County, and Other. Only the expenditure assumed to be financed by County funds is included in the operating and capital costs. It also assumes that 50% of the school capital expenditure and 100% of the road capital expenditure are financed by debt.

County and State Funding: This case differs from the County Funding approach by adding the expenses associated with State funding of the school operating budget. It is more realistic as the obligation to fund the schools falls on the County and not the State. Roads are also financed by debt.

100% PAYGO using county's per-student costs & roads: This case differs from the County Funding approach by assuming 100% of the capital expenditure for schools and roads is funded by PAYGO. The per-student costs for school construction are the County's numbers. That this approach uses the County's funding approach for school operating costs.

100% PAYGO using updated per-student costs & roads: This assumes that 100% of the capital expenditure for schools and roads is funded by PAYGO. However, the per-student school construction cost used in this approach is much higher as the County's approach underestimates this cost significantly. The school operating cost remains unchanged from the County Funding approach.

Fiscally Conservative Approach: The fiscally conservative approach is the most responsible approach. First, it takes into account the County and State funding needed to operate the school system. Second, it uses 100% PAYGO to fund road and school construction. Third, the per-student school construction cost is much more accurate than used by the County.

Note that in all the cases, the amount of debt the County continues to service using the General Fund is not included.

Analysis Results

The County's six-page fiscal impact analysis result (found here) incorporates the FY2023 adopted operating and capital budget. The two sets of primary inputs are residential unit types and job types by planning area. The HoCo By Design fiscal impact methodology and assumptions outputs an 18-year projection of revenues and expenses for various categories of sources of funds and obligations, respectively.

Since the County’s results do not provide the expected annual net revenue, an independent analysis was performed (found here) to reproduce the County’s results. Further a sensitivity analysis was performed to evaluate the impact of changes in assumptions. Several scenarios are generated including a fiscally conservative scenario.

The County’s fiscal impact analysis concludes that new growth would generate more revenues than costs for services and infrastructure. Several assumptions are made in the entire analysis about residential development, which also revealed county budgeting practices, to enable this assertion. Before delving in to the other assumptions, a financial analyst would look at the results table and immediately notice something strange. The Average Annual column of the results table takes the simple average of the annual revenue and expenditure results and takes their difference to obtain $58,872,000 in net average revenues. Next, the first year net revenue is shown to be $29,791,000.

Two Basic Major Issues

As a long-term Pro Forma financial projection these two results are problematic:

First, the simple average of an 18-year financial projection ignores the time value of money. For example, in year 18, the County's result says revenues of $304,732,000 are expected. This is a meaningless number. For instance, the expected revenue in year 18 when discounted to 2023 at the 20-year U.S. Treasury Bond yield of 4.118% is approximately $147.6 million.

Second, the projected net revenue in year one needs to be placed in perspective. The FY2023 General Fund revenues total $1.29 Billion. That means, the projected net revenue is just 2.35% of the total revenue not counting the rest of the budget. No amount of prediction can provide this level of precision, especially with the level of variability encountered when dealing with amount of assumptions used.

This leads to the next point.

Additional Major Issues

The Fiscal Analysis Does not Employ Representative Historical Values

Howard County's budget and financial history did not start in 2023. The biggest predictor of the future is the past. In statistics one data point is useless. In order to gain a degree of statistical confidence, more data points are needed. This is a standard practice in the task of predictive statistics. For example, drug trials rely on multiple trial participants of various ages and backgrounds (if not they should). If the test subject for a drug trial is one 25-year old white male, the drug company cannot say with any degree of confidence whether or not the drug works. If the trial has multiple 25-year old white males, the degree of confidence improves, but it does not say how it will affect 25-year old white females, black males, or black females. It also does not say how it will affect older males, younger males, older females, younger females, children, or any other demographic group.

This simple analogy can be applied to the County's fiscal impact analysis. The FY2023 budget represents one economic period in time. It does not explain Howard County's economic and fiscal experience in 2022, 2021, 2020, and beyond. Each annual budget is a reflection of the set of zoning and land-use decisions that took place in the previous years. Therefore, it only makes sense to use a representative number with some sort of associated variability to capture the range of possible outcomes. In other words, an average of several fiscal years and a standard deviation.

The next set of analyses pertain to specific sections of the County's fiscal impact results.

HCPSS Operating and Capital Expenditure

Per Student Operating Costs

The County's analysis uses a per-student operating cost of $11,341.32 using the FY23 enrollment of 59,367 students as its basis. Meanwhile, the County states that the total per-student operating cost is $17,353, which includes State and other funding. The County states that only the local funding is taken into account because this is what is accounted for in the County General Fund. There are two issues with this assumption and methodology.

First, the per-student cost uses FY23 enrollment numbers while the per-student debt financing cost (discussed later) uses 57,325 as a basis. Using a higher number of students here depresses the per-student cost.

Second, why is only the County General Fund taken into account? The assumption the County takes should reflect what would happen if the County is required to cover shortfalls due to decline in state funding. This leads to the suggestion that the County should use total per-student cost; not partial per-student cost. Every additional residential home with a student incurs an operational demand of $17,353.

Per Student Capital Costs

The school capital budget calculation and assumptions also face similar challenges. State aid to build schools is part of the cost to build a school and the primary responsibility of building a school falls on the County; not on the state. Therefore, excluding it assumes that the dollars to cover the balance after the County's general fund contribution will always be available and this is obviously not true. If the state does not award the aid, the County defers much needed school construction.

Next, there are serious flaws on how the County calculates the per-student construction costs. For example the County estimates high schools cost roughly $130 million based on Guilford Park High School. Excluding state funding it estimates a general fund obligation of $77,686,000. However, this seriously underestimates the cost. For example, while the school construction project at Hammond High School added 200 and cost $106,554,000 it was also a renovation project. Recognizing this distinction, there is still a significant cost to the County to add seats to an existing school than to build a new one.

Nobody benefits from under counting or not recognizing this impact. In fact, it provides a much better and accurate picture. After all, the need to add 200 seats are due to new residential development. As a result, the per-student cost to build high school is closer to $127,315 without recognizing state aid. Even assuming state aid, the actual cost is closer to $77,000.

Another major glaring flaw of the methodology is the assumption that only 50% of the capital costs for schools will be PayGo and the remaining 50% would come from debt financing over 20 years. The County justifies this approach by claiming that a portion of the school surcharge collect pays old debt.

The County says:

Even though the majority of school surcharge revenues are paying debt service from previous projects, new development still pays this surcharge so should get credit to offset school capacity costs incurred.

This is a rather shocking and false claim. If someone pays of their previous month's credit card debt using this month's paycheck, while incurring new debt for this month to cover living expenses, did they actually pay off all their debt? No.

Just in 2023 alone, the county spent $48,747,588 in debt service for school capital costs. There is no scenario under which significant amounts of county dollars are allocated to finance debt incurred for school construction to accommodate previous years' new residential development that would be lead to a conclusion that "residential development pays for itself."

Systemic Renovations Assumptions

The County's accumulated systemic renovation costs are over $500 million. The analysis assumes a per-student basis of $580.90 based on a total spending of $33,300,000. There is no indication that this amount will help reduce the accumulated deferred maintenance over the next 18 years. In fact it will most-assuredly not, given that every year more facilities need upkeep.

Student Generation Rates

Student generation rates have long been contentious. The County assumes that a student from a new residential development is no longer tracked as such for planning and amortization purposes after the first year. In other words, after the first year, a new student is no longer from a new residential development. Further, in the past, the County has shared data that large amount of the fifty plus communities create resales. Retirees sell their home to a young family and move to a fifty plus community. The new student from that resale is clearly tied to the new construction and should be treated accordingly.

Road Expenditure

The County's analysis of road expenditure has several flaws.

First it adjusts to inflation total annual historical spending on roads. As mentioned earlier, neglecting to employ some sort of a net-present value methodology to the 18-year projection is one of the analysis flows in projecting future revenue and expenditure. So it is interesting that this section applies the time value of money to adjust old spending to 2022 dollars.

Next, the County uses an incorrect 2010 County population. In 2010 the County population was 288,627. Not 247,842. Using this lower value reduces the per-capita spending that is used to project future expenditure. While the analysis uses $1,388 per capita and jobs created, the real number is $2,087 per capita and jobs. This would underestimate the spending by over 33%.

Finally, the fiscal analysis says "it is assumed that 100% of capital costs for new roads and related capacity expanding improvements will be debt-financed over 20 years..." It is assumed that there is no PayGo funding for these future roads given that all future Road Excise Tax revenues are currently being used to pay back existing Road Excise Bonds." In other words, residential development does not pay for itself.

Conclusion

As of FY2022, the County’s debt is standing at $2.1 billion. If the county actually paid for the cost of schools and roads as well as other infrastructure, between now and 2040 the county debt would balloon by anywhere from $450 to $500 million reaching $2.5 billion. Clearly this general plan has serious flaws.

We need a general plan that will help us get out this giant, massive hole we are in due to terrible planning and land-use decisions. We have public schools with massive disparities in achievement gaps that need to be addressed.

The top five underperforming elementary schools are predominantly attended by children of color and children from low-income families. We need all the resources at our disposal to ensure that children who need the attention have access to schools with small class sizes.

We need a high school in Elkridge. It is incredibly shocking to hear that in a meeting with the county administration the Elkridge community was given a choice between a community center and a high school. We need to protect taxpayer dollars by advancing a responsible general plan to ensure an equitable allocation of resources.