Are We Underestimating Funding Shortfalls?

The Toronto Sun reports, Feds have $65B pension funding shortfall:

The federal government has underestimated its employee pension obligations, exposing taxpayers to a $65 billion shortfall, according to a new report released Thursday by the C.D. Howe Institute.

Using fair-value accounting, the measure used in the private sector and based on solvency, the think tank calculated Ottawa’s net pension obligation stands at nearly $208 billion. That’s $65 billion more than reported in the public accounts.

The government lists its unfunded liabilities in the country’s national debt at $143 billion.

Taxpayers could be on the hook to back-fill the funding gap, the report by Alexandre Laurin and William Robson said.

On top of that, large exposure to public sector pensions could fuel fears of sovereign default driving up the cost of borrowing.

"The larger-than-reported gap between federal pension promises in these plans and the assets that back them is a problem, both for federal employees and for taxpayers,” the pair said.

But Canada is not alone. European and U.S. governments share the problem. The United Kingdom for instance is facing a $1.8 trillion shortfall and the U.S. has $3 trillion.

The difference, according to the report, is that heavily indebted countries such as the U.K. have made reducing these obligations a top priority.

Most federal employees in this country have what’s known as defined-benefit pensions, the most stable of all benefit plans since contributions are fixed. And higher-income public servants often qualify for special retirement compensation arrangements.

Backing promises to public service workers, the RCMP and Canadian Forces would require contribution rates of 35%, 41% and 42% of pay respectively, the report says citing the chief actuary.

Actual contributions to these plans today are 19%, 22% and 21% respectively. On average, two-thirds of costs are borne by the government.

Leave it to the C.D. Howe Institute -- a conservative think tank that vigorously defends private sector interests -- to grossly exaggerate the true state of public pension shortfalls. I went over their study, and it's based entirely on fair-value reporting of Ottawa's pension obligations:

More importantly, the federal government arrives at the $201.4 billion liability figure by discounting its accrued benefit obligations using notional interest rates. One of these – a legacy from before April 2000, when federal pensions were completely unfunded, and Ottawa needed a benchmark to track its accumulating obligations – is the interest rate on 20-year federal bonds for obligations arising from service before then. The other is the expected return, currently about 4.2 percent in real terms, on fund assets for benefits earned since April 2000. Neither rate reflects current reality.

Deferred compensation is akin to a loan from employees to the government – in this case, a loan indexed to inflation and backed by taxpayers. For that reason, the best interest rate for discounting the obligation is the yield on federal real return bonds (RRBs). At the RRB rate on March 31, 2010 – 1.56 percent – liabilities for 2009/10 would have totalled about $255 billion, as shown in the second column of Table 1.

The final entry in the first column of Table 1, “unamortized estimation adjustments,” is the portion of changes in asset values and liability estimates, using the government’s accounting, not yet reflected in the Public Accounts. The fair value column contains no such entry, because fair-value accounting recognizes all such changes immediately.

The government’s net pension obligation under the fair-value approach thus stands at almost $208 billion – some $65 billion larger than reported in the Public Accounts.

This raises the net public debt by an equivalent amount. And, because the gap between reported and fair-value pension obligations has grown over time (Figure 1), these adjustments also change the annual budget balances. Since 2001/02, the Public Accounts show the cumulative budget balances to be almost exactly zero, with surpluses and deficits offsetting each other. The fair-value approach to pensions, by contrast, shows a cumulative deficit over that period of $72 billion. In 2009/10 alone, the annual deficit would have been not the $55 billion reported, but $63 billion.

I have a problem using the yield on RRBs to project future liabilities. Moreover, fair-value reporting is sketchy, especially when you consider where we are in the cycle and that private investments (private equity, real estate and infrastructure) are not going to be valued fairly using this method. The other thing to bear in mind is that most pension funds do not have to sell their private investments any time soon, which is another reason why relying solely on fair-value reporting grossly distorts the true funding state of public pensions.

***Feedback***

Bernard Dussault, the former Chief Actuary of Canada, shared these comments with me:

The investment yield assumptions used for the projections of pension liabilities in this report do not appear relevant to me, as it is deemed that all categories (including equities) of assets will on the long run yield the same low rate as bonds. This artificially and unduly increases the value of pension liabilities.

The real rate of return of 4.3% assumed for equity investments by the Chief Actuary in his triennial actuarial evaluations of the cost of federal pensions might well be considered somewhat optimistic (e.g. I think it should not exceed 4%) but it would not be reasonable to assume for equities a real rate of return of only 2.75%.

And Patty Ducharme, National Executive Vice-President of the Public Service Alliance of Canada, put out this letter:

The C.D. Howe Institute report described in this article is another attempt to rob hundreds of thousands of hard working Canadians of their retirement security.

By re-releasing their own skewed figures, the C.D. Howe Institute is attempting to strike fear into the minds of Canadians by basing the whole premise of their projections on a doomsday scenario whereby Canada’s federal government ceases to exist.

The Institute's estimate of the “fair-value” costs of the federal public pension liabilities assumes that the federal government will one day cease operations in the same way as a private corporation going bankrupt.

But the latest actuarial report tabled in Parliament in November 2009, in fact, shows that the federal public service plan is adequately funded and is running a surplus – a far cry from the $65-billion liability that C.D. Howe is ringing false alarms over. This surplus is due in part to the large contribution by employees. Contribution rates of public service employees to the federal public service pension plan have been increasing substantially since 2006.

The viability of the pension plan is not in question. On the contrary, recent reports from the Public Sector Pension Investment Board indicate that the market investments of federal public service pension funds are generating significant returns to more than cover future pension plan liabilities.

The authors of the C.D. Howe report should go back and review their research.