Public pensions are a ticking timebomb

The other day I was talking with my mother-in-law about retirement planning, tax reform, and government spending. She’s in the enviable position of having a pension which covers all of her living expenses. She knows she will be able to pay her bills and put food on the table for the rest of her life, regardless of how the stock market performs and regardless of what happens to interest rates.

Our discussion essentially revolved around my assertion that pensions are an unfair tax on future generations. A pension is a promise that benefits will be paid in the future for services provided today. In effect, current taxpayers are forcing future taxpayers to pay for current services.

My mother-in-law disagreed. Her belief is that today’s taxpayers are funding all of the costs for employing public sector workers today. The workers’ salary and benefits are paid out of current tax revenue, and money is set aside to pay for future pension benefits.

It’s the idea that sufficient money is set aside today to pay for future benefits that I want to talk about today. While correct in theory, the reality is that the structure of public pension programs is such that it’s almost a given that future taxpayers will be forced to fund at least some of the benefits for today’s public sector employees.

How a public pension works

For this discussion we’ll focus on state public pensions, such as those given to teachers, firefighters, and police officers. At its most basic level a public pension is when government employees (politicians) promise other government employees (public sector workers like teachers, firefighters, and police officers) benefits that will be paid in the future.

The recipients of public pensions have generous pension programs in part because they have powerful unions. The unions support certain political candidates and in return the candidates agree to protect and/or improve the pensions for the unions.

In most states, these promised pension benefits are protected by the state constitution. There is no way, short of a potential bankruptcy, to stop paying or even reduce the pension benefits that have been earned. To reiterate – no matter how insolvent a state’s finances, the state’s constitution typically guarantees pension payments. Thus, pensions must be paid before schools can be funded, roads can be built, or any other public service can be delivered.

The way that pension funding is supposed to work is that money is set aside today to pay for future benefits. To make the numbers simple, let’s say that a employee starts working at age 25 and is promised a pension worth $1M at age 65 (in 40 years). NOTE: this is called a “defined benefit” plan. The employee is being promised a specific benefit, regardless of how the underlying investments perform in the meantime.

And again, to make things simple, let’s just ignore inflation for this simple exercise.

In real life the public sector employer (state or local government) would contribute money to the pension program every year, but to keep things simple let’s just say that the employer will make a single, one-time contribution today to fund the pension program’s future benefits.

How much money does the employer need to contribute? Well, it depends on the expected yearly return. Let me repeat that, because it’s really the crux of everything we are talking about (and the problem with pension programs) – the amount of money that must be set aside today depends entirely on the expected yearly return from now until the pension benefits must be delivered.

For the sake of this exercise, let’s assume an 8% yearly return. Using a spreadsheet we can quickly calculate that an employer needs to contribute $49,713.41 today to have a $1M balance in 40 years.

Pretty simple so far, right?

Now here’s the big question – what happens if the employer doesn’t have $49,713.41 today to fund the future pension? Well, the employer has a few options:

  • Contribute less than $49,713.41 today and plan on making up the difference in the future, or
  • Assume a higher average yearly return

Politicians don’t like to take the first approach. Intentionally underfunding public pensions is bad politics. If you underfund a pension for too long you’re going to lose the support of the teachers and firefighters unions that helped get you elected in the first place (and who you’ll need to get reelected).  It’s also worth noting that this option is economically difficult. If you don’t have the money to contribute today, why would you think that next year you’ll have the money to make next year’s contributions PLUS the shortfall from this year?

Instead, it’s easier to just assume a higher return. The higher the expected return the lower the amount of money that needs to be contributed today for the plan to be “fully funded”.


Assuming higher returns in the future is essentially a magic wand that current politicians can use to promise generous future benefits without needing to worry about funding them today.

Small changes in the expected return can yield big changes when long time periods are considered. Changing the expected return from 8% to 9% per year reduces the amount of money needed today from $49,713.41 to $34,702.96. That’s a 31% reduction in the amount of funding that must be made today!

So if you want to promise $1M in the future but only have $34,702.96 available to fund a pension, you can just assume 9% returns in the future and suddenly the pension plan is “fully funded”.

This works the other way as well. If you assume an 8% return but actually get a 7% return you’ll end up being significantly short in 40 years.

For the first few years the difference is small. By the end of year 5 there’s only a 3.65% difference, and by the end of year 10 it’s only an 8% difference. These shortfalls are easily explained away as market fluctuations, temporary setbacks, etc.

By year 23 the public pension plan is in big trouble. The plan is supposed to have $270k (assuming an 8% return) but it only has $220k (because the plan only earned 7%/year). The plan is $50k behind schedule, which is more than the amount of the entire original contribution!

And even if the plan actually had $270k it still wouldn’t be enough. $270k is what’s needed in year 23 if the plan earns 8%/year. How much would the plan actually need to have in year 23 to have $1M in year 40 at 7%/year return? The plan would need to have $316,574.39, which means the plan has a $96,324.02 shortfall.

The pension plan is screwed.

So what happens in year 40, when the state is bound by its constitution to pay the $1M in benefits? The then-current taxpayers have to come up with $304,269.78 to pay the remainder of the promised benefits. They must pay these benefits before schools are funded, roads are built, or any other public services are funded.

Pension plans in real life

In the above example it was easy to see that the plan was underfunded. We were only using a single person who had their pension funded with a single payment. In real life you have money flowing into the plan every month (as employers make payments to fund current employees’ future pensions benefits) and money flowing out to pay the pensions of current retirees. The balance of the pension plan fluctuates depending on the underlying investments, yearly contributions, the ratio of current employees to retirees, etc.

Let’s look at how this works in real life. For this discussion I’ll be examining CalPERS (California Public Employees’ Retirement System) because:

  1. I live in California
  2. CalPERS is pretty middle of the road in terms of funding levels
  3. CalPERS is one of the largest public pension funds

According to the CalPERS 2014 Annual Report (page 127) CalPERS assumes 2.75% annual inflation and 7.5% annual returns.

If Calpers earns even a little less than 7.5% average returns the plan will end up being dramatically underfunded. Why?

Because public pension plans have no end date.

They are designed to be permanent. Public sector employees are being hired TODAY that will need to receive benefits in 40 years and those benefits will need to last for a typical 20+ year retirement. Small differences in assumed return vs. actual return will be massively magnified over those 60 years.

So how has CalPERS done over the last decade? Have they been on track to hit their assumed 7.5% return? We’ve been in the midst of the second longest bull market in history, so 7.5% annual returns should be easy to achieve, right?

Well, from 2007 to 2017 CalPERS has averaged a mere 5.1% return (remember, they assume a 7.5% return). We saw what a mere 1% difference in returns can do to a pension plan – what do you think a 2.4% underperformance will do?

The good news is that CalPERS (and other pension plans) are realizing that their assumptions might have been a bit too rosy. CalPERS has reduced their expected returns from 7.5% to 7%. Why? Well, from 1996 to 2016 CalPERS only earned 7%/year. That’s a 20-year period of underperformance. Apparently CalPERS finally decided to have their expectations match their historical performance.

CalPERS is not alone. Across the country, public pension plans have reduced their expected returns from an average of 8% returns in 2001 to 7.6% in 2015. Expected returns from individual plans range from 5.5% to 8.5%. However, since 2001, the average annual returns for public pension plans have only averaged 5.1%!

This consistent underperformance has had effects across all state pension plans. Overall public pension funding levels dropped from 75.6% in 2014 to 74.5% in 2015 to 71.1% in 2016. New Jersey has the lowest funding level at just 30.9%. Other interesting states are New York at 94.5% funding and Washington DC at a full 100% of funding.

And remember, ANY plan will be “fully funded” if you assume a high enough future return. If I’m managing a pension plan for you and you have a promised benefit of $1M in 40 years, I can put $1 in an account today and claim that the account is fully funded. How? I just need to assume an annual return of 41.25%! If I assume a return of 45% I can say that the pension plan is OVERFUNDED and I can either pull money out or defer future contributions.

The overall public pension funding level of 75.6% in 2014 is based on overly optimistic projections of future returns. If more realistic expectations were used the public pension funds would have significantly lower funding levels.

And if these pension plans are underfunded, who is going to pay the future benefits?

Public pensions are a tax on future taxpayers

The real crime here is that we are forcing future taxpayers to pay for services we are enjoying today. After all, if pension benefits are protected by a state’s constitution then those pension payments MUST be made. If the pension plan is only 50% funded then future benefits will be 50% paid by the pension plan and the other 50% will need to be paid by future taxpayers.

The sad reality is that public pension plans inevitably lead to this result. It’s just too easy for current politicians to make promises that future generations will be responsible for. Who cares if the promises are reasonable (or even feasible)? No politician wants to be seen as anti-teacher, anti-firefighter, or anti-police. It’s so much easier to just make promises and paper over any funding deficiencies by assuming unrealistically high future returns.

The way to fix this is to move away from defined benefit public pension plans and instead switch to defined contribution plans (like virtually everybody else in the private sector has). The funding for employees’ retirement plans is done during their period of employment. No future obligation is created so there’s no way to defer responsibility to future taxpayers.

Note that I’ve only been talking about PUBLIC pensions. I have no issue with private pensions (pension programs offered by GM, IBM, or other corporations to their employees). In private pension programs the current management of GM is promising future pension benefits that the future GM will have to pay. If the pension plan is overly generous and/or not sustainable then GM might be forced to declare bankruptcy, but it’s not going to affect future taxpayers (short of another bailout, of course).

If GM declares bankruptcy then the PBGC (Pension Benefit Guarantee Corporation) will step in and make at least minimum payments to participants in GM’s pension plan. This is something of a self-correcting model because if the employee’s union insists on unrealistically high pension benefits then the company will be bankrupt and only minimal payments from the PBGC will be available. This removes the incentive that public employee unions have to push for ever higher pension benefits, regardless of ability to pay. Private pension participants have at least some incentive to be “reasonable”.

I should also point out that almost all of my family members are or will receive a public pension:

  • Father – Community college president
  • Mother – Jr. high and high school English teacher
  • Sister – Guidance counselor at high school
  • Step-mother – Community college administrator
  • Step-father – University professor
  • Sister-in-law – Public sector employee
  • Brother-in-law – High school teacher
  • Wife’s cousin – Elementary school teacher

I certainly appreciate the advantages of public pensions to the people who receive them. I just think the public pension model is inherently flawed and the current implementation is going to lead to a lot of pain down the road.


So what’s the takeaway here?

First, if you have a public pension, you need to plan for the possibility that you won’t receive all of your promised benefits.  Only Washington DC and Wisconsin have fully funded pension plans (and again, these funding levels are inflated because the plans are STILL assuming overly optimistic future returns). Check your state’s laws to see how “guaranteed” your future benefits are. And remember, even if your pension benefits are guaranteed by the state constitution, constitutions can be changed and/or Federal law could be changed to allow states to declare bankruptcy and wipe out pension obligations.

Second, whether or not you have a pension, you should use very conservative return numbers when trying to project your future net worth. When I read magazines or blogs most people tend to assume somewhere around 10% returns. According to a report by Vanguard, 10% annual returns are really only achievable if your asset allocation is approximately 100% equities. An allocation of 80%/20% equities/bonds results in a 9.5% annual return. A more conservative 60%/40% allocation (the typical prescription for 41-year-old investor like me) should return 8.7% annually. The difference between 10% and 8.7% returns is large over long periods of time. It’s probably safe to match your allocation to Vanguard’s report and then plan for slightly lower returns (8.25% returns at a 60/40 mix, for example). It’s far better to save more than needed and be able to retire early than it is to not save enough money and work longer than expected.

Third, as a taxpayer, you should assume that future taxes will be higher than current taxes. This means you should max out Roth accounts before non-Roth accounts (higher future taxes make tax-deferred accounts less valuable and make tax-exempt Roth accounts more valuable). Strongly consider higher allocations to tax-exempt investments like state municipal bonds. If you’re young, create a saving and investment plan to reach your goals with assumption that you’ll have less money to invest in the future (as more money will be taken up by higher future taxes). You should also assume that capital gains and dividend taxes will be higher, as it’s possible that States will pass taxes on both types of gains to help pay for future benefits.

Fourth, you might want to avoid living in states with lower levels of pension funding, as they are most likely to have the biggest tax increases in the future. Citizens of New Jersey (with a funding level of just 30.9%) are going to be on the hook for enormous public pension payouts. If you’re trying to decide between living in New Jersey or Wisconsin I’d factor future tax liability into your decision.


Questions for you

Are you a public sector employee? Do you have a pension, and if so, are you worried about the solvency of your pension fund?

Regardless of whether you receive a pension or not, do you think public pensions are a good idea?



9 thoughts on “Public pensions are a ticking timebomb

  1. I am a firefighter in one of the largest departments on the east coast, and pretty much every one of my coworkers is banking 100% on their pension payout upon retirement. I am hopeful, but not entirely dependent upon it for survival after my career is over. Fortunately, the population where I work continues to rise and the area is growing significantly, which means the taxpayer base continues to grow and revenues have continued to rise.

    On the flip side of all of your points, the unions are not sitting by and soaking the population we serve dry. We continue to increase employee pension contributions (mine currently sits at 10% of my total salary, new hires are paying over 12%), lower salary payouts upon retirement (mine was reduced to 50% of pay from 75% and even 85% for the real old timers), and future hires will have their retirement pushed from 20 years to 25 at the earliest.

    Things will definitely have to change for the future, but it won’t all fall on the taxpayers. The unions and the politicians will have to continue to work together and find common ground in order to make it work for the taxpayers and for the people who are earning those pensions. Firefighters, police officers, and teachers are the backbone of these communities and should be taken care of for the service they provide during their careers.

    1. To be clear, I’m not saying that public sector benefits are unfair, and I’m not saying that the unions are soaking the population. After all, my parents rely on their public sector pensions to fund their entire retirements.

      The problem I have is that a defined benefit program is making a promise about future benefits, regardless of what happens between now and some point in the future.

      Having new employees pay more is helpful, but all that really does is change today’s funding. It doesn’t address the root problem of unrealistic assumptions about future returns. To use the example in my post, I can promise $1M in 40 years an only contribute $1 to a pension today if I assume 40%+ annual returns. It doesn’t really matter if the $1 is funded by taxpayers or employees or 50/50. Either way a promise has been made that can’t be fulfilled.

      And I certainly believe that public sector employees provide valuable services. I just don’t think it’s a good idea for ANYBODY to make a promise to perform a service in 40 years NO MATTER WHAT. That’s effectively what a public pension is – a promise to provide future benefits regardless of whatever happens in the intervening years. That’s just a bad idea.

    2. I also think that it’s fundamentally unfair to the employees to make unrealistic promises to them. As you said, all of your colleagues are relying on their pension, and the reality is that the money might not be there to provide the pension benefits they have been promised. If they were offered smaller but realistic pensions then they’d know they need to save and invest today. By making false promises we are really screwing our public sector employees doubly.

  2. As someone that deposits a CalPERS check every month from a job I left 23 years ago, I’m going to tell you that most of the people I worked with then would not be able to manage a defined contribution plan. They would invest it inappropriately, borrow from it, and do everything wrong. And that’s some of the better educated, more fiscally conservative people with accounting and real estate backgrounds and jobs. In the end, the taxpayers will have to support these people in some fashion. I would rather the pension system support them with investments than the general taxpaying population with then current revenues.

    I think more realistic assumptions about returns going forward and a period for these pension funds to catch up to reasonable funding levels is the best way out of this mess. Contributions from the employers to make up the underfunding of the past and both parties to fund the future pension contributions seem like the best compromises. Yes, the taxpayers will have to kick in, but they are already subsidizing healthcare and will probably have to pay for some of the Social Security promises as well.

    1. The management of a defined contribution plan is pretty easily solved. After all, virtually the entire private sector manages it just fine. You could just put people into a retirement date fund, or a 75/25 stock/bond plan, or something else reasonable. Or you could just have a defined benefit plan where a trustee manages the money (essentially what CalPERS does now) and then just say that at retirement the employee’s money would be used to purchase an annuity (which is essentially what a pension plan is).

      The problem with changing the assumptions to something more reasonable is that each downward adjustment of assumed future returns automatically reduces the current funding level. That is, assuming 7% returns instead of 7.5% returns might change the funding level from 80% to 70% since you need a lot more money today at 7% annual returns vs 8% returns to hit some future benefit value.

      But I think you’re right that changes will need to be made. Unfortunately, those changes are almost certainly going to result in taxpayers today making up for underfunding in the past, and taxpayers in the future making up for underfunding today.

      It’s a total mess.

  3. The problem with purchasing an annuity on the date of retirement is that paper asset values cannot be predicted. What if the equity markets take a 50 percent hit the month before you retire? Your retirement is gone. It’s the power of large groups comprised of contributors and annuitants that make pension plans work. The large oscillations of the market can be smoothed away if the group is large enough.

    Buying an annuity adds another level of fees and relies on one company to provide the benefit. The large pension plans have huge diversified asset portfolios and incoming contributions for stability. What happens if the company that offers the annuity gets caught like a couple almost did in 2008?

    People want certainty. They want a guaranteed income starting on a date certain. If you do not offer them that, they will not participate in the plan. If you force them to participate and the asset markets tank when it’s time to retire, there will be torches and pitchforks. No politician wants that to happen and they will not be inclined to vote for such a plan.

    I really think some pain in the near term to shore up the pension plans is the way to solve this. No more Hail Mary passes like CalPERS has been trying for the last decade or two. Suck it up and fix the system over the next 20 years, and everyone takes a small hit.

    1. The good news is that I think we’re saying the same thing – there is substantial risk in projecting the future values of current investments. Your concern is that there might not be enough money in investments at retirement to support a public sector worker. My concern is that there might not be enough money at retirement to prevent future taxpayers from being required to pay substantial amounts of money to pay for promised benefits. Why should taxpayers get stuck with the bill if the stock market crashes right before a given employee retires? The fundamental problem is that public sector employees are being promised a future benefit REGARDLESS of what happens between now and retirement.

      Your concerns about certainty and annuities are all valid, but remember, these are all the exact problems that the entire public sector must wrestle with. I don’t see any reason that public sector workers should receive substantially better benefits than private sector workers.

      Note that my solution (convert all public sector pensions from defined benefit to defined contribution plans) would have simple fixes to the problems you mention. The easiest plan would be to continue to have CalPERS manage all investments. Employees cannot withdraw funds before retirement. The big difference is that your investment returns would be tracked, and at retirement you’d have some specific chunk of money. That could then be annuitized or withdrawn in one big chunk. If you want to have something approximating a more traditional pension then CalPERS could just tell you what your monthly payment would be based on your balance at retirement.

Leave a Reply

Your email address will not be published. Required fields are marked *