There is no question that the promises made by state government pension plans are a major challenge to the future financial health of U.S. states. Many states pensions are already dramatically under-funded. A recent study suggests that the aggregate under-funding in state budgets, including pensions, is over $4 trillion. A 2010 analysis, the definitive research to date, finds that the unfunded state pension promises amount to $2.5 trillion.
What do these kinds of numbers really mean in practical terms? Joshua Rauh, author of the second study above, finds that each American household will need to pay an average of an additional $1,398 in tax each year, forever, to allow the states to keep their pension promises to state employees. The amount varies from state to state. Five states will require each household to contribute more than $2000 per year. I say that the tax increase is ‘forever’ although the study actually looks out only 30 years into the future. For most taxpayers, this is effectively a permanent tax increase.
If pensions end up under-funded, there are not all that many solutions. First, the states can raise taxes to make up the shortfall. Second, the states can raise the contributions that current employees make into the retirement system and raise the retirement age. Third, in the worst case, the state may need to pursue some way to reduce the pensions that they pay to their current retirees.
Estimates of how badly state pension plans are under-funded vary. Differences in projections for the funding adequacy of pension funding come from two main sources. The first is the assumed rate of return that the pensions can generate on their investments. The second factor is the way that the pension accounting deals with investment risk. Rauh, with co-author Robert Novy-Marx, argue that the current accounting for state pensions allows the plans to plan for high expected returns (the average for all states is around 8%) from a range of risky investments and then to treat these investments as essentially risk-free. The problem with assuming that a pension will definitely receive an 8% rate of return on its investments is that there is no investment that can provide this guarantee. Rauh and Novy-Marx argue that this approach dramatically understates the true needs of pension funds.
This problem with public pension accounting is, interestingly, something that many personal finance practitioners recognized and dealt with years ago. When you invest in a risky asset, you must account for that risk. Risky assets such as stocks have higher expected returns than a risk-free investment but may also go through extended periods of under-performance. Investors need to save more when you account for the impact of risk. How do you measure this?
Using computer models that create (simulate) thousands of possible future returns from the market, for example, you can try to make sure that the worst outcomes are survivable. This process is achieved through what is called a Monte Carlo Simulation (MCS) that uses a computer to ‘roll the dice’ to generate a range of future outcomes for a portfolio.
The name “Monte Carlo” derives from the famous gambling center. This approach, first developed for simulating problems in statistical physics, is a standard of practice in portfolio management. In mid-2008, before the market crash really hit, I did a simple analysis of the range of outcomes that an investor could reasonably expect over a ten-year holding period with an S&P500 index fund. My results suggested that a fairly optimistic view of the future (8.3% expected return for the S&P500) still resulted in a 1-in-20 chance of about a 10% cumulative loss for a ten-year holding period. And that result was in nominal dollars (e.g. not adjusted for inflation). If you don’t account for this range of possible future returns, you don’t need to plan for the possible future in which stocks deliver negative returns—you can just plan for the 8.3% return every year.
While there is legitimate disagreement among a range of experts, I find Rauh and Novy-Marx providing the most compelling analysis. If their estimates are correct, the traditional pension plans of this country face an enormous crisis. The under-funding of U.S. public pensions will ultimately be paid for by either taxpayers or public sector workers who will be forced to accept less than they have been promised in retirement. Pensioners are somewhat like bondholders in that they have been promised a specific stream of future payments by a state or municipality. There is, however, considerable ambiguity in how pensioners will fare relative to bondholders in the event that a state or municipality goes bankrupt.
This issue should be of paramount concern to everyone, whether or not they have a public pension. For private-sector workers and retirees, the risk is that the shortfall in pensions will be made up in the form of higher taxes. In addition, it is quite possible that that public services will be curtailed as states devote an ever-larger fraction of their budgets to pensions for current retirees. For public-sector workers, the implications are fairly obvious. Current workers can expect to be asked to bear a substantial portion on the burden in the form of higher contributions to their retirement plans and later retirement dates.
For now, it is in everyone’s interest to see a bright light shone on this issue. We don’t want to end up trying to deal with these issues at the last possible moment, when pension plans are actually on the brink of insolvency.