Shifting New Jersey Lottery to Pensions is a Gamble

Gambling on behalf of pensions may no longer just be a pastime of New Jersey’s pension fund investment managers. In May, New Jersey Gov. Chris Christie’s final budget address included a proposal that would transfer the state lottery to the state’s pension fund, banking on future gambling revenues to fund retirement benefits.

Specifically, Gov. Christie proposed to dedicate future lottery revenues to “eligible pension plans,” — a move that the governor claims would help improve the solvency of the state’s very underfunded retirement plans. Under the Christie plan, the lottery would become an asset owned by Teachers’ Pension and Annuity Fund, the Public Employees’ Retirement System, and the Police and Firemen’s Retirement System. The future revenues of the state lottery would be dedicated to these funds — primarily the teachers’ fund — as an additional source of revenue to pay for pension benefits.

On its face, this would be a boon to the New Jersey pension system. Gov. Christie argues that his plan would reduce the unfunded liability by $13 billion, improving the state pension system’s funded ratio from 49% to 64%. And the state treasurer estimates the lottery revenues would amount to $37 billion over 30 years.

Yet, pulling a few strands on this idea shows that it’s more palliative than panacea.

The state pension funds would get the benefit of the asset transfer, but take on the risk of over-counting those future revenue dollars. Lottery revenues have declined somewhat in recent years, and banking on an overestimated revenue stream from the lottery would be just as troublesome as the state’s pattern of overestimating investment returns.

As our colleague Len Gilroy said in a recent interview, “there is a risk if in the future, for some reason, lottery usage would decline.” And that risk is counting future revenues as an asset today, which in turn would reduce the amount of unfunded liabilities, which in turn would reduce unfunded liability amortization payments.

If future experience leads to a decline in lottery usage, and thereby the lottery’s asset value, then the current $13 billion estimated value will have been too high, and additional contributions will be needed in the future to make up for that.

And worse, counting those chickens too early will have meant prior unfunded liability amortization payments were too small, thus turning the lottery transfer into a cause for even further underfunding than currently exists. (This is a similar problem to states that use ad valorem taxes — based on the value of the good, e.g. sales taxes — to collect revenue from marijuana or alcohol sales. If sales or prices decline enough, the state misses out on expected revenue.)

At the core of this proposal is simply an accounting shuffle. Consider that even if the lottery generates $37 billion over 30 years that this is $37 billion the state will have to back fill from other funding sources. Lottery revenues of about $1 billion annually are currently dedicated to things like education and public health services. Switching their dedicated funding target over to pensions would likely lead to a scramble for around $1 billion in general fund revenue to provide for those education and public health services lottery is currently funding. This, in turn, could mean legislators decide to prioritize these programs over required contributions to the pension system. If this leads to a one-to-one tradeoff of money for the pension system, the state will be in the same place. Part of Christie’s pitch for the proposal is that it would be “budget neutral” for this exact reason. But standing still isn’t an option for New Jersey.

On the flip side, the state could simply promise to pay $37 billion a year in general fund revenue over the next 30 years — and this would not generate a $13 billion asset transfer. In either scenario ($37 billion in lottery revenue direct to pensions, or that lottery revenue going to the general fund which is then paid into the pension funds) the same amount of money would be going into the pension fund, but on a different accounting basis. That one of these magically increases the funded status of the pension funds by treating the lottery as an asset is a slight of hand reflecting many of the reasons that pensions have become so underfunded nationally in the first place.

There’s also a legal question. The New Jersey Constitution authorizes the state to run a lottery, “when the entire net proceeds of any such lottery shall be for State institutions and State aid for education.” If a court were to find that a pension system isn’t an “institution,” this would eliminate the possibility of using these funds for any system other than teachers’ pensions. This is unlikely to happen (if no other part of the NJ pension system qualifies, the Teachers’ Pension Annuity Fund almost certainly does), but it could throw a wrench into the proposal even after it clears the legislature. The state’s attorney general has weighed in, saying the proposal is constitutional, though “[g]iven that this is a novel transaction, and in the light of uncertainties associated with litigation in general, we are unable to say with absolute certainty how a reviewing court would decide a legal challenge.”

Ultimately, even if the lottery does very well in the future, it is unlikely that Gov. Christie’s proposal would fully address the problems facing the solvency of the state’s pension funds any more than simply promising to pay 100% of actuarially determined contributions and paying those out of general fund revenue first every year. Dedicating revenue is great — on the margin this is probably more stable a funding source than the year-to-year budget debates — but that doesn’t mean this proposal is any kind of substantive reform. Real solutions in New Jersey will have to mean structural changes to funding policy and sustainable benefit design reforms.