Item 2: Pension Report — original pdf

Backup

Sustainability of Austin’s Pension and Post-Employment Benefit Plans November 19, 2025 Austin has lost a lot of money. The City of Austin’s financial position is recorded in the Annual Comprehensive Financial Report (CoA ACFR). The latest report from 2024 shows that “total governmental activities net position” was positive $662,050,000 in 2015 and is negative $541,244,000 in 2024. (Table 1, page 277.) The report says, “The deficit in governmental unrestricted net position is largely due to the net pension liability of $2.3 billion and other postemployment benefits (OPEB) liability of $1.9 billion.” (Page 4) In plain English, the government of Austin has lost more than 1 billion dollars over the last 9 years due to pensions and healthcare for its retirees. This report for the Economic Prosperity Commission’s Long-term Working Group investigates this problem, what caused it, and the City of Austin’s current response to it. It proposes additional changes to prevent its repeating. Since OPEB and pensions are separate topics, the report is divided into 2 sections, covering each separately. Members of the Working Group Michael Nahas, author Aditi Joshi, commenter Zain Pleuthner, commenter Shakeel Rashed, commenter Expertise Michael Nahas, the author of this report, has a Masters of Arts in Economics from UT-Austin. His thesis was on the relationship of foreign bond prices to currency exchange rates. He worked in finance, predicting stock prices and futures prices for an options trading firm. He also wrote bond trading software for 2 other firms. He is not an accountant. He is not an actuary. He is not an expert in pensions. Members of the Working Group talked to: ● Ed Van Eenoo, Chief Financial Officer of the City of Austin ● Christopher Hanson, Executive Director of COAERS ● Keith Brainard, Vice Chair of the Texas Pension Review Board ● Joshua Rauh, Professor of Finance at Stanford Graduate School of Business Executive Summary Austin is 6th in the country in per-capita employee retirement debt. This is the OPEB liability plus the pension liability. Other Post-Employment Benefits (OPEB) ● OPEB is a big problem: $2.5 billion risk-free net liability ● The biggest cost of OPEB is medical insurance. Insurance is getting more expensive and people are living longer, so costs are increasing. ● Austin has no savings to pay for OPEB. ● Austin is not required by law nor contract to pay for OPEB. It can stop at any time. In a financial crunch, our retirees should expect worse medical insurance. In a ● bankruptcy, our retirees should expect to lose their medical insurance. ● Austin should implement a gradual fix. New hires should be covered by a funded defined-contribution plan. Pensions ● Pensions are a big problem: $6 or $7 billion risk-free net liability. ● COAERS is the largest pension. Twice the financial size of police plus firefighters’. ● Austin “borrowed” around $100M per year for 24 years from our retirees in COAERS. ● From 2002 to 2010, Austin was on a path to never pay off the “debt” to COAERS. ● Austin’s “debt” is $3.5 billion with a 6.9% variable interest rate. (For reference, Austin’s ● GO Bonds have a fixed interest rate below 4.3%.) In 2021-2025, Austin helped write state laws to fix the problem. These laws require “ADEC”, which changes Austin’s payment to the pension funds from year to year. ● At least one pension has a “maximum employer contribution”. It is not specified what happens if that is exceeded. ● Austin will continue to “borrow” more from COAERS until 2031. ● We will spend around $111M per year for 28 years paying off the “debt” to COAERS. ● If predictions hold, we will have paid roughly $4.6 billion in “interest” to COAERS when it is paid off. (That number is adjusted for inflation and is in 2025 dollars.) ● Future losses will be paid off over 21 to 25 years. ● If predictions hold, every future net loss of $1 will be paid back by $1.50 (adjusted for inflation). ● A 2020 report listed 4 causes of the “borrowing”. Some, arguably, still exist. ● Austin needs reports that show the full range of possible future scenarios. ● City Council needs to closely monitor the fix enacted in 2021-2025. ● City Council should pay off liabilities quicker to avoid interest payments. ● City Council should know the effect on pension liabilities before signing contracts. ● Taxpayers should be informed of the City's financial losses every year. Notes Before Reading The audience of this report is the residents of Austin. Where possible, technical terms and equations have been avoided. Clarity was chosen over precision. Thus, you will see ‘Austin “owes” $3.5 billion to its future retirees.’ rather than ‘The sum of the net pension liability for City of Austin’s three pensions is $3.5 billion.’. Some technical terms are unavoidable. To help, there is a glossary at the end of this document. The numbers in this report are large. While reading, it is handy to remember that the population of the City of Austin is about 1 million persons. Thus, $1 million means $1 per person and $1 billion means $1,000 per person. Thus, ‘$3.5 billion’ means $3,500 per person. Keep in mind that “person” is not just working adults; it includes children, homemakers, retirees, etc. Other Post-Employment Benefits (OPEB) What is the problem? A short answer to the question “What is the problem with OPEB?” is that the City of Austin’s future retirees expect healthcare and the City has not set aside any money for it. To understand the issue, some background information is necessary. I will start with a short overview of OPEB in general. Then, I’ll explain the state of Austin’s OPEB. Lastly, I’ll explain what should be done to fix the problem. Overview of OPEBs After working for years at an employer, a retiree may get “other post-employment benefits” (OPEB). “Other” refers to things other than a pension. For the City of Austin, other benefits include “access to medical, dental, and vision insurance for the retiree and the retiree’s family and $1,000 of life insurance on the retiree only.” The most expensive of these benefits is medical insurance. OPEB can be paid for using an investment fund. Money is put into the fund while the employee is working and taken out during retirement. For the fund to work correctly, the expenses must be predicted correctly. For example, predicting how long each worker will live and predicting the cost of medical insurance. With those predictions, the right amount of money can be invested in the fund. In the future, that money and the interest or dividends of the investment will be used to pay the benefit. Austin’s OPEB Austin does not have an OPEB fund. The CoA ACFR 2024 says: “Allocation of City funds to pay postemployment benefits other than pensions is determined on an annual basis by the City Council as part of the budget approval process on a pay-as-you-go basis.” It goes on to say: “The City is under no obligation to pay any portion of the cost of other postemployment benefits for retirees or their dependents.” In contrast to these words, the CoA ACFR 2024 expects these benefits to be awarded each year and contains an estimate of the City’s current liability for future costs. While not legally bound to pay OPEB, the City has habitually paid them. In accounting, a habitual non-required payment is called a “constructive obligation” and is included as a liability. This is why the CoA ACFR includes an estimate of the OPEB liability. The CoA ACFR 2024 contains a short explanation of how it predicts the lifespans of our current employees and the cost of future healthcare for them. No one in the working group has the actuarial expertise to comment on those predictions. Using those predictions, the future payments needed by the city can be determined. The CoA ACFR 2024 calculates the current liability from those future payments using a discount rate. The discount rate is usually chosen to be a low-risk bond. For the discount rate, the City used 3.26%, which is the interest rate for the class of bonds containing the City of Austin’s 20-year General Obligation bonds (“GO bonds”). Using this discount rate, the total OPEB liability for Dec. 31, 2023 was $3,245,417,000. I find that liability to be too conservative (too large). Municipal bonds, like Austin’s GO Bonds, get a lower interest rate because they are tax exempt. It is acceptable to use a US Treasury bond, because it is considered a risk-free bond. The 20-year US Treasury bond recently had a rate of 4.875% and that higher rate results in a lower liability. Using the sensitivity analysis from the CoA ACFR 2024, I extrapolated the liability with a discount rate of 4.875% and got $2.5 billion dollars, which is 23% lower. What Caused the Problem? For the OPEB, the cause of the problem is that we have never put any money aside for medical care, dental care and vision care. (Or, alternatively, that we keep paying for it even though we are not legally obligated to.) The most likely explanation for not putting money aside is that, historically, medical care was cheap. The cost of a few visits by a doctor in 1970 was small compared to pensions. But medicine has progressed with new tests, new treatments, new medicines and has gotten expensive. Since 1970, the per-capita spending on healthcare has increased by 6 times, adjusted for inflation. Also, retirees are living longer past retirement. American lifespans have increased about 9 years since 1970. Thus, OPEB has gotten expensive as we must pay more for healthcare and must cover retirees for longer. What Has Been Done to Fix the Problem? As far as I can tell, the City has done very little. Voluntarily, it continues to pay medical care and other benefits to retirees. It neither makes those benefits permanent nor announces an end to the benefits. Because OPEB is not a legal obligation, it cannot cause the City to go bankrupt. And, because of that, its liability does not affect our bond rating. However, because it is not a legal obligation, there is a risk of a sudden end to the benefits to our retirees. When the City’s budget becomes tight in the future, payments for OPEB can be stopped and the money spent on other priorities. In the case of the City’s bankruptcy, all benefits would go away. This is what happened when Stockton, CA, went bankrupt. What Else Should Austin Be Doing? It is not good to have a multi-billion dollar expectation. Especially one where payments keep growing. It is not viable to establish a fund (a legal trust) for the current OPEB benefit for all employees and fully fund it. First, the City does not have $2.5 billion to put in. Second, contracting for the $2.5 billion liability would lower the City’s credit rating and increase its borrowing costs. Lastly, it would bind the City to pay higher and higher costs for OPEB as medical expenses increase and people live longer. Instead, I recommend a gradual approach: ● RECOMMENDATION HIGH IMPORTANCE: The City should establish a fund (a legal trust) to pay OPEB for all employees hired after the fund was created. The future lifespans of retirees and future costs of medical care are difficult to predict and will cause difficulties in determining how much money to put into the fund for each employee. Thus, I recommend a fixed percentage of wages be put into the fund to pay for OPEB benefits for each employees’ retirement. This is called a defined-contribution plan. For existing employees, they should remain under the implied defined-benefit plan, funded by the City on a pay-as-you-go basis. Existing employees will benefit from the shrinking size of the unfunded OPEB liability — the risk of a sudden end to their OPEB will be decreased over time. Existing employees will also benefit from higher wages, which would not be available if City’s budget tightens in order to fund future OPEBs. Over time, as new employees roll onto the new plan and former employees roll off the old plan, more of the City’s employees will be on the defined-contribution plan. The controlled OPEB costs will help the City afford the payments to the defined-contribution fund while it pays the pay-as-you-go expenses for retirees. I also recommend: ● RECOMMENDATION HIGH IMPORTANCE: The City should continue to explore the cheapest way to provide medical care to existing retirees. The Affordable Care Act, known as ObamaCare, subsidizes medical insurance for some participants on the healthcare exchanges. Chicago has lessened its OPEB liability by having retirees contract their own health insurance. Essentially, allowing federal subsidies to pay for city liabilities. The City could give money to retirees for premiums and still come out ahead. We should examine this approach. Pensions What is the problem? A short answer to the question “What is the problem with pensions?” is that the City of Austin owes pension payments to future retirees and has not set aside enough money to do so. It is, effectively, borrowing from its pension funds. It helps to think of the situation as borrowing. When the City doesn’t invest enough today, it loses the interest and dividends that those missing dollars would have earned over time. To catch up later, the City has to put in more money, like paying interest on a loan it made to itself. To understand the details of this will take a few pages of explanation. I will start with a short overview of pensions in general. Then, I’ll explain the state of Austin’s 3 pensions. Lastly, I’ll review what Austin has done recently to fix the problem and make suggestions about improvements to it. Overview of Pensions Pensions are a benefit provided by employers to employees, where the employer will pay employees after they retire. An example of a simple pension is: employees are allowed to retire at age 62 and receive 75% of their highest salary until they die. Pensions are paid using an investment fund. Money is put into the fund when the employee is working and withdrawn when the employee is retired. The money put into the fund by the employer is called the “employer contribution”. Often, employees are allowed or required to put a portion of their income into the investment fund. That money is an “employee contribution”. For a pension to work financially, the pension’s income must exceed its expenses. The income depends on (1) employer contributions, (2) employee contributions, and (3) interest or dividends generated by the investment fund. The expenses depend on (4) the highest salary of each worker and (5) how long each worker will live. Parts 3, 4, and 5 must be predicted when the employee is working. If the predictions are wrong, there will not be enough money in the fund and the employer must pay the retiree directly: paying money when the employee is not working. Pensions are often called “defined benefit” plans, because the benefits are set at the time of employment. An example of a defined benefit is: “employees are allowed to retire at age 62 and receive 75% of their highest salary until they die”. It is the “defined benefit” aspect that requires prediction, because the cost of those benefits are unknown at time of employment. Austin’s Pensions Austin has 3 pension funds. They are for the police, firefighters, and all other employees: ● City of Austin Police Officers’ Retirement and Pension Fund ● Firefighters’ Relief and Retirement Fund of Austin, Texas ● City of Austin Employees’ Retirement and Pension Fund The last fund, the one for “all other employees”, is known as known as “COAERS” For scale, here are the number of active employees and retirees in 2024: Employees Retirees Police 1,541 1,592 Firefighters COAERS 1,249 1,068 11,197 12,082 COAERS is the largest pension, covering 4 times as many employees/retirees as the other plans combined. Each of the 3 pensions has an investment fund that has bought investments, like stocks and bonds. The money to buy these assets came from contributions in the past. Below is the current price of each funds’ assets and a breakdown of the kinds of assets. Police Firefighters COAERS Market Value of Assets on Dec. 31, 2024 $1,081,418,113 $1,155,244,470 $3,512,867,909 Stocks (“equities”) 54% Bonds (“fixed income”) 20% Other (Level 1 or 2) Other (Level 3: “significant unobservable inputs”) 0% 26% 44% 30% 0% 26% 59% 17% 24% 0% The last category, labeled "significant unobservable inputs” in the pension reports, is difficult to price. There are no open markets for it, nor similar open markets where the price could be inferred. This category includes investments like private equity, private credit, some real estate, and natural resources. (COAERS was not able to invest in these in the past, but recent changes will allow it in the future.) Because this category of investments is difficult to price, the market value of assets is an estimate. How much do we “owe” the pensions? Ideally, when an employee works, the City should put enough money into the pension fund to cover the retirement benefits they earned that year. The amount only needs to be a fraction of the actual benefit, because the fund’s investments should grow over time. Calculating the exact amount that should be in the fund is difficult because the amount of money needed depends on a number of predictions. To calculate how much we should have in the pension fund, predictions are made on: ● Lifespans of employees and retirees ● Wage increases for employees ● Other employee behavior (Will they quit City of Austin? Marry? Etc.) ● Future investment returns For example, the predicted return on the investments for the three pensions are: Predicted investment rate of return Police 7.25% Firefighters COAERS 7.30% 6.75% Because these are predictions, there are multiple ways to think about how much we should have in the pension fund. Later, I’ll discuss how to measure a worst-case estimate. For now, I’ll focus on the “best guess” estimate, which is the average estimate. Because of some special circumstances, the best guess of how much we should have in the funds can be found in the 2024 pension reports under the heading “total pension liability”. If we subtract what we have now (market value of assets plus a few other things) from the total pension liability, we have our best guess of how much we “owe” to the pensions. That is the “net pension liability”. Total Pension Liability on Dec. 31, 2024 Net Pension Liability on Dec. 31, 2024 Police Firefighters COAERS $1,878,457,525 $1,514,813,506 $5,957,378,923 $795,679,941 $349,466,268 $2,386,653,425 The net pension liability for all three funds together is $3,531,799,634. That is our best guess of what Austin “owes” to the retirement funds. Based on the predictions. How important are the predictions? Net pension liability is a “best guess”, given the predictions, but how important are those predictions? This section will look at just one prediction: the predicted return from the investments. You will see that Austin may owe more and should use a more prudent, financially conservative estimate of its liability. The pension reports acknowledge that there is risk not expressed in the net pension liability. The pension reports include the calculation of a liability called the “low-default-risk obligation measure” (LDROM). When we take the LDROM liability and subtract the current market value of assets for all three pensions, we get $5.3 billion dollars, or 50% higher than the net pension liability. But the LDROM liability is not a worst-case estimate of the liability. It is calculated using the price of a bond that still has risk. It uses “AA-rated” bonds, which are low-risk but still have a higher chance of default than than “AAA-rated bonds”. Corporate pensions measure their liability using AA-rated bonds, but most financial economists compute liabilities using the price of risk-free bonds. (Anyone interested in the details should look up “discount rate” or “discount factor”.) Thus, the LDROM underestimates the potential liability. A better way to calculate the worst-case liability uses the price of a rock-solid investment, like the U.S. Treasury Bond. It is the closest thing we have to a risk-free bond. I used the “sensitivity” values in the pension reports to extrapolate the net pension liability with a discount rate of 4.875%, which is the recent rate for a 20-year U.S. Treasury bond. The result was $6.1 billion dollars, which is 72% higher. Stanford University’s Joshua Rauh and Oliver Giesecke wrote a paper “Trends in State and Local Pension Funds”, which mentions the problem with liabilities and discount rates. Their website https://publicpension.stanford.edu/apps/page_local has data for Austin. It does not have data for 2024, but calculates the market value of the pension liability in 2023 at $7.06 billion. It is important to remember that net pension liability is our best guess at how much we owe. But it contains predictions and, if our predictions are wrong, we could end up owing much much more. As much as twice more, just for the prediction of investment returns. Implications Because Austin has not invested enough money in the pension funds, we are essentially borrowing from our retirees. Each year that we don’t buy financial assets, we have to make up for that the next year by paying the interest those assets would have generated. Which means we are essentially borrowing billions of dollars at the high interest rate of 6.75% to 7.30%. And that is a variable interest rate — it could be higher some years. “Borrowing” is an appropriate term and the pension reports themselves call the missed earnings “interest”. The image below is from the 2024 police pension report, page 108, which calculates the “total experience gain or loss”. The previous year’s liability multiplied by the previous year’s predicted investment return is labeled “interest”. (UAAL stands for “unfunded actuarial accrued liability” and is a smoothed version of net pension liability used to determine contributions.) In the 2024 firefighter’s pension report, it is on page 119. In the 2024 COAERS pension report, it is on page 111. Thus, it is correct to say that we are “borrowing” from our pensions at a 6.75% to 7.30% variable interest rate. For reference, Austin was able to borrow for 20-years in September 2025 with General Obligation bonds (GO bonds) that charged a fixed interest rate of 4.29%. So, “borrowing” from pensions is both riskier and more expensive. What Caused the Problem? The problems with the pension started about the year 2000. That was the burst of the “Internet bubble”, which caused both a drop in stock prices and a tight City budget. Since 2000, we’ve “borrowed” an average of $100,000,000 per year from COAERS. This is visible in the graph below. This graph shows what we “owed” COAERS over time. The data is the “unfunded actuarial accrued liability” (UAAL), which was more available than the net pension liability. The UAAL is similar to the net pension liability, but includes smoothing because it is used for planning investments into the fund. Because the values are derived from predictions, the values can shift if predictions change. And predictions have changed over the years. The predicted return on investments was decreased in 2015, 2019, and 2021. Thus, earlier values on the graph are likely lower than they would be if graphed using the pensions’ current predictions. City Council passed increases to the City’s contribution to the pension funds at various times from 2005 to 2013. These did not fix the problem. In fact, a City report says the COAERS had an “infinite amortization period from 2002 through 2010”, which means planned contributions were not enough to ever pay the pension. The police pension had an infinite amortization as late as 2018. As to the cause of the problem, there was a presentation named “Pension Analysis Report” for the City’s Audit and Finance Committee given on June 3, 2020. On page 7, it said the problems with the pension were: ● Being too optimistic about investment returns (“investment returns below actuarial predictions”) ● Being too optimistic about other predictions (“Revised actuarial assumptions to adjust for this experience”) ● Putting in too little money (“Fixed annual funding falling short of actuarially determined contributions”) ● Not even trying to pay off the debt (““Tread water” funding dynamic”) What Has Been Done to Fix the Problem? Around 2019, the City of Austin started to fix all the pensions. The fix has 2 parts. One part was to create a payment plan called the “legacy liability” to pay the current “debt”. The other part was to move from a fixed contribution each year, to a variable one. Thus, when the City “owes” more money to the pension, the City puts in more money. The technical term for this is an “actuarial-determined employee contribution” or ADEC. Implementing the legacy liability payment plan and ADEC required renegotiations with the employees and changes to state laws covering Austin’s pensions. It was implemented by the following Texas laws: ● Police Pension: HB 4368 in 2021 ● COAERS Pension: SB 1444 in 2023 ● Firefighter Pension: HB 2802 in 2025 These laws contain additional changes. Some increased the employee contribution rate. The police and firefighter pensions created new “tiers”, where new employees will have less benefits or require longer service before qualifying for retirement. But the big change in all of them is to pay off the existing debt with a legacy liability payment plan and avoid new long-term liabilities with the ADEC, which increases the City’s contribution if more money is “owed”. For all pensions, the legacy liability payment plan will pay off the current net pension liability over 30 years. Let’s look in detail at COAERS, since it is the largest pension. After a 2 year phase-in, the legacy liability payments start at $111,160,870 and increase by 3% per year. Payroll is expected to rise by 4.5% to 6.0% per year, due to inflation and raises, so these payments should become a smaller and smaller percentage of the pension contribution over time. The final payment is in 2053. This graph shows what we will “owe” the COAERS pension, if all predictions hold. For dates up to 2024, it is the UAAL. From 2025 on, the data comes from “Remaining Legacy Liability” on page 90 of the COAERS 2024 report. It is difficult to see in the graph, but the liability goes up until peaking in 2031. This is because legacy liability payments are below the “interest” on the “debt” until 2032. Thus, Austin will continue to “borrow” more from our retirees in COAERS for the next 6 years. This graph is an estimate of the “interest” payments to COAERS by year. This is calculated by multiplying the UAAL and Legacy Liability by the expected investment return rate. To compute the sum of all “interest” payments, inflation was calculated using CPI-U before 2025 and using the 10-year breakeven inflation rate of 2.28% after 2025. The sum total of interest payments for COAERS, in 2025 dollars, is $4.7 billion dollars. In addition to the legacy liability payment plan, the fix has the ADEC, which adjusts the employer contribution based on the current net pension liability. It has a complicated calculation, so I’ll only describe some parts of it. If COAERS’s assets don’t achieve the predicted 6.75% return, the loss is smoothed over 5 years. Then, the difference between the “actuarial asset value” and the expected asset value is packaged as a new “liability layer” and paid off similarly to the legacy liability. That is, it is paid off in equal annual payments over 20 years or until 2053, whichever is longer. There is a lag time to the employer contribution calculations. For example, a shortfall in 2026 will only be seen after the year ends. The liability layer payments will be calculated in 2027 and start being paid in 2028. Thus, a loss may be smoothed over 5 years, and, after each year, there is a lag year and then 20 years of payments. So, losses may take 26 years to pay off. For comparison, most corporate pensions pay off losses over 7 years, after a 1-year lag. I computed equal payments for a $100 loss paid off over 20 years at a 6.75% interest rate with a 1-year lag. The payments were $9.88 per year. If inflation is 2.28%, which is the current 10-year breakeven inflation rate, the interest payments total $50.28 for “borrowing” $100. Since smoothing lengthens the time of payments, actual values for COAERS would be larger. For comparison, if the loss was paid off in 7 years with 1 lag year, the interest payments would total $21.63, or 57% less. The COAERS report says that if the ADEC causes the employer contribution “to exceed the maximum contribution rate then the [employee contribution] will increase up to an additional 2% of payroll to cover the shortfall. If a 2% increase in the [employee contribution] is insufficient then the System will enter into discussions with the City to come up with a plan to restore the funding of the System to a satisfactory level.” This statement sounds like Austin’s fix is incomplete. I have not seen a calculation of the probability that the maximum contribution rate is exceeded. What Else Should Austin Be Doing? I have a large number of recommendations. Most are directly targeted at the long-term financial stability of the pensions. The recommendations have various importance and I will state the importance with the recommendation. Financial Stability The City’s latest changes to the pensions are complex. And previous changes did not fix the problem. It is important to know how the City’s changes will perform. I do not think the sensitivity analysis included in the pension reports is sufficient. ● RECOMMENDATION HIGH IMPORTANCE: The City Budget’s “Taxpayer Impact Statement” page should include the per-ratepayer change in the City’s “total governmental activities net position” for the previous year. This is probably the most read page of the City Budget. Austinites should know if the City is losing money, because their taxes will have to be raised or services cut. ● RECOMMENDATION HIGH IMPORTANCE: The new pensions rules should be simulated under random expected conditions (for example, shuffled historical data) and varying assumptions. The output of those simulations should include the range of values for net pension liabilities in 2053. It should also include the range of employer contributions and the probability that the employer’s “maximum contribution rate” is exceeded at any point. The impulse response should be computed for a once-in-a-decade and a once-in-a-century loss. (A report containing some of this information may already exist; I requested a copy but have not received it. I did receive slides from a presentation from August 25, 2023, but those do not show any of these recommended values.) ● RECOMMENDATION HIGH IMPORTANCE: If simulations show that there is a significant probability of exceeding the employer’s “maximum contribution rate”, City Council should work to specify what happens. It is better to negotiate that now than when a crisis happens. ● RECOMMENDATION HIGH IMPORTANCE: The City’s reports should include a range of values for the pension liability based on the distribution of assumed returns. Meketa’s report on the assumed future returns of each asset class does not have a single value, but a range of values. It would be good to include in the reports the range of values for the net pension liability based on the different ranges of returns for asset classes, especially in the Austin Comprehensive Financial Report and the City Budget. This will remind readers that the single liability number is the result of a prediction. The pension reports currently include a sensitivity analysis, but that is missing context. It shows the effect of a 1% misprediction in each direction. But mispredictions are likely to be much larger. Meketa puts the “likely ranges” for US stocks as +4% to -4%. Thus, COAERS reported sensitivity of $0.6 billion to a 1% change might actually mean we should expect shifts on the scale of $2.4 billion. ● RECOMMENDATION HIGH IMPORTANCE: When negotiating contracts, City Council should be informed about the effect on the pension liabilities. The pension liabilities depend on predictions, such as the future wages of employees, because the amount paid out depends on the employee’s salary. If a contract raises wages faster than predicted, it will have an immediate effect on the pension liabilities. City Council needs this information when negotiating a contract. According to City staff, it has not had this. ● RECOMMENDATION Medium Importance: Each pension fund should track the effects of mispredicted assumptions since the ADEC was implemented. If anything should cause worse-case pension liabilities to increase, we want to identify it quickly. One of the likely causes for pension liabilities to increase is bad predictions. Each pension should report the sum of increased liabilities since the ADEC was implemented that were caused by each class of prediction. The prediction classes are: investment returns, actuarial predictions (lifespans), employee behavior (quitting, etc.), and others. If the most recent changes to the pensions do not tame increasing net pension liability, we will easily know the cause and may detect problems sooner. I believe this data is calculated annually for each pension, but not summed over all years for all pensions. ● RECOMMENDATION Medium Importance: Pension funds should report correlations with well-known risks that cross multiple investments. These are risks like inflation, interest rates, S&P 500, VIX, individual currencies, oil, and geographic proximity. The City should also measure these correlations with its other assets and incomes, such as sales tax. It is prudent to limit exposure to any risk across a portfolio of investments. This is already being done by individual pensions for some risks. I did not see indications that it is done across all of Austin’s assets. ● RECOMMENDATION Medium Importance: Adjust retirement age or years of service for longer lifespans. Humans are living longer. In 1941, when COAERS was established, the average American lifespan was 64 years. It is now 79. For COAERS, the retirement age is 62. And, for the police and firefighters pensions, members can retire at any age with 23 or 25 years of service. The City of Austin is paying more and more for a longer and longer retirement, but only receiving the same number of years of work. That means pension expenses will keep climbing as a percentage of wages. Austin should adopt a formula now to increase the retirement age or years of service as American lifespans get longer. I suggest recalculating the retirement ages for new employees each decade. This may be set by state law, but Austin has recommended changes to state law many times. ● RECOMMENDATION Medium Importance: When pensions predict future wages, they should use the wages that the City has already contracted for. The liability calculations require a prediction of future wages, but we already know some of those future wages. For example, the City signed a contract to raise police wages by 28% over the next 5 years. Those contracted wage increases should be used when calculating the pension’s liability. ● RECOMMENDATION Medium Importance: Voluntarily pay-off liability layers with losses over 7 years. Currently, they are paid off over a 20-year period. That is very long. It means after a loss, the City is slow to pay back the deficit and it loses the benefits of the return on investments. Thus, we will pay more “interest” over time and residents will pay more taxes. Most corporate pensions amortize losses over 7 years. Once-in-a-century losses might need to be paid off over a 20-year period, but that can be handled on a case-by-case basis. ● RECOMMENDATION Low Importance: Voluntarily pay off the legacy liability faster than is in the payment plan. Delaying repayment of the net pension liability causes a lot of large “interest” payments. We should avoid those. This item is only marked “low importance” because of the political difficulty of implementing this solution. ● RECOMMENDATION Low Importance: Explore replacing the defined-benefit pension plans with defined-contribution plans. They are easier to manage. Defined-benefit plans are difficult to run well, as shown by Austin running up multiple billion dollars in “debt” and pensions’ role in city bankruptcies like Stockton, CA, and Detroit, MI. Other cities have changed their plans. Houston’s solution to its pension problems included a clause that reduces the benefit when investment returns are low. Most new corporate retirement plans are defined-contribution plans, like 401(k)s. This item is only marked “low importance” because of the political difficulty of implementing the solution. ● RECOMMENDATION Low importance: Do not base benefits on the highest salary earned. This policy makes it difficult to predict the correct amount of money to put into the investment account to fund benefits. It also results in perverse incentives for employees, who can receive a windfall in benefits from a single promotion. This risk is so well known that it has a name: “pension spiking”. It has happened to many cities, including Phoenix. Most corporate pensions do not include this policy. Those pensions use “career average pay”, which makes liabilities more predictable and is not subject to abuse. City staff said that overtime is not included in Austin’s pension benefit calculations and that avoids the worst versions of pension spiking. This item is only marked “low importance” because of the political difficulty of implementing this solution. Prudence Accounting has a “Doctrine of Prudence”, also known as the “Convention of Conservatism”. The Doctrine of Prudence is about financial safety. It says to book future losses immediately and book potential gains when they happen. ● RECOMMENDATION Medium Importance: Request that the Government Accounting Standards Board (GASB) require a prudent, financially conservative discount rate and use it for certain liabilities. Austin follows the rules set by GASB. State law and bond markets require it, so Austin cannot set its own accounting rules. Thus, if we want Austin to use a prudent, financially conservative discount rate for the liability included in the ACFR, City Council must request changes from the GASB. The request should ask that the LDROM liability calculation should be based on U.S. Treasury bonds or another risk-free interest rate. The request should also allow Austin’s ACFR to use that liability, not the net pension liability, as a prudent, conservative value to calculate the City’s “total governmental activities net position”. This is the advice of Stanford University’s Joshua Rauh and Oliver Giesecke. ● RECOMMENDATION Medium Importance: Pension funds should report external sources for their predicted investment return rates and other predictions. COAERS is the only one that gave its source for the investment return rates, Meketa. ● RECOMMENDATION Low Importance: Use the same predictions of inflation rate, predicted return rates for asset classes, etc. across all pensions. If the pension liability is meant to be a measure for the City’s financial state, it is inconsistent to use different predictions for the different pensions. Conclusion This ends my report. I believe I got a rough understanding of how the OPEB and pensions operate, what some of the major problems were, and was able to make good recommendations. I am disappointed that I was unable to diagnose a specific cause that allowed Austin to run up $3.5 billion in “borrowing” from the pensions. The City does follow state law, Texas’s Pension Review Board decisions, and GASB regulations. But those were insufficient to restrain “borrowing” and City leadership, with its peak at City Council, did not implement financial discipline. Even now, I suspect that the pensions’ predictions are too optimistic and future problems are likely. Given the long time-horizons, the predictions involved, and that voters are not knowledgeable enough about defined-benefit plans to enforce discipline on City Council, I wish that Austin did not have its defined-benefit pensions. I hope that a defined-contribution plan for OPEB will avoid problems in the future. Glossary: ACFR: Acronym for “Annual Comprehensive Financial Report”. Each year there is one for the City of Austin and one for each of the three pensions. ADEC: See “Actuarial Determined Employer Contribution” Actuarial Determined Employer Contribution: Rather than a fixed employer contribution each year, the amount can change and is calculated using an actuarial formula. Asset: A financial asset is a stock, bond, cash, or other investment that is valuable. CoA: Acronym for “City of Austin” COAERS: City of Austin Employees’ Retirement System. It is the pension covering all of Austin’s employees, except police and firefighters. It does cover EMS. COLA: Acronym for “cost of living adjustment” Contribution: The money put into the pension’s investment fund. Cost of living adjustment: An increase in the payout of the pension to retirees. This might be used to account for unexpected inflation between the work year and retirement years. Defined-benefit plan: A retirement plan where the benefits are specified. For example, a pension where the employee’s benefits are stated in the agreement. (For comparison, see “Defined-contribution plan”.) Defined-contribution plan: A retirement plan where the employer puts in a fixed amount each year that the employee is employed. Common ones are known as IRA (Individual Retirement Accounts) and 401k. (For comparison, see “Defined-benefit plan”.) Discount Factor: See Discount Rate. Discount Rate: An interest rate used to convert prices at times in the future or past into a price “now”. A 5% discount rate would convert a payment of $1.05 in a year into a “now” value of $1. For pensions, it is used to convert payments due to retirees in the future into a value for this year’s accounting. Employer contribution: The part of the contribution paid by the employer. (See “contribution”) Employee contribution: The part of the contribution paid by the employee. (See “contribution”) Government Accounting Standards Board: An organization that sets the accounting rules for cities, like the City of Austin. General Obligation bond: This is the most common way for the City of Austin to raise money from financial markets. It is a bond, an IOU. “General obligation” means that there is not a specific tax or source of revenue devoted to paying back the bond. If these bonds are used to pay for infrastructure, they are tax-exempt. GO bond: Short for a “general obligation bond”. Liability: Money owed to someone. Market Value of Assets: The price of all stocks, bonds, and other investments if they were sold right now. Net pension liability: A number reflecting how much money needs to be added to the pension’s investment fund so that all future benefits will be paid. It depends on a lot of predictions about future investment returns, lifespans of employees, etc.. It is the total pension liability minus the market value of assets and a few other small things. OPEB: An acronym for “other post-employment benefits”. That is, benefits besides the pension or defined-contribution plan. The biggest expense in OPEB is health insurance. But it might include dental, vision, life insurance, etc.. Total governmental activities net position: An accounting measure of all that the government-side of City of Austin owns. It does not include the utilities, like Austin Water, Austin Power, etc.. UAAL: An acronym for “Unfunded actuarial accrued liability” Unfunded actuarial accrued liability: It is a smoothed value of net pension liability, used for planning future contributions to the pension fund. U.S. Treasury Bond: A loan to the federal government of the USA. Vesting: The determination of when an employee qualifies for a benefit. E.g., Employees vest after 5 years of employment”. Links: For those with printed copies, these are the URLs linked above: CoA ACFR 2024 https://www.austintexas.gov/sites/default/files/files/Finance/Financial%20Transparency/CAFR/F Y2024-City-of-Austin-ACFR.pdf Austin 6th in per-capita pension debt https://reason.org/commentary/public-pension-debt-rankings-for-state-and-local-governments/ Police Pension Reports https://www.ausprs.org/publications/annual-financial-reports/ Firefighter Pension Reports https://www.afrfund.org/publications/annual-financial-reports/ COAERS Pension Reports https://www.coaers.org/annual-financial-reports Chicago used ObamaCare to lessen OPEB liabilities https://billmanzi.com/2013/05/15/shedding-opeb-through-obamacare/ Stanford University's website on municipal pensions https://publicpension.stanford.edu/apps/page_local Austin Monitor's page from 2002 on budget discussions https://austinmonitor.com/stories/2002/01/economic-trends-must-change-to-prevent-bigger-short fall-director-of-financial-services-john-stephens-and-budget-officer-rudy-garza-painted-a-grim-ec onomic-picture-for-members-yesterday-early-budge/ 2019 City Report on pensions https://services.austintexas.gov/edims/document.cfm?id=331704 "Pension Analysis Report” for the City’s Audit and Finance Committee given on June 3, 2020 https://services.austintexas.gov/edims/document.cfm?id=341544 Houston's solution to its recent pension problems https://www.houstontx.gov/pensions/ Pension Spiking Wikipedia page https://en.wikipedia.org/wiki/Pension_spiking Pension Spiking story for Phoenix on Ballotpedia https://ballotpedia.org/City_of_Phoenix_Initiative_to_Amend_Pension_Plan,_Proposition_487_( November_2014) Doctrine of Prudence Wikipedia page https://en.wikipedia.org/wiki/Convention_of_conservatism