Skip to main content

The Liberal government's proposed changes to the tax rules are raising a lot of flak from physicians and other high-paid professionals. By taking away income splitting with one's spouse and children, the government says it is trying to promote fairness among all taxpayers.

If you're not directly benefiting from income splitting, the proposed changes probably sound reasonable. The only question is whether the tax revenue is worth the potential political fallout. I understand the change might generate $250-million in extra tax revenue but when compared to a budget deficit of $28-billion, that doesn't sound like a lot.

If the federal government really wants to make a dent in the deficit and promote fairness, it should consider what can be done in its own backyard. There are nearly half a million active and retired civil servants who participate in the federal Public Service Pension Plan (PSPP for short). If any program is in need of an overhaul, it is the PSPP.

For subscribers: Morneau vs. business: How Ottawa created its tax troubles

I contend that the PSPP should be converted from a defined-benefit pension plan to a target-benefit plan. The basic difference is that employer contributions to a target-benefit plan are fixed, which means the benefits themselves can be variable. In a defined-benefit plan, the amount of benefit is unconditionally guaranteed, which is possible only if contribution rates are variable. It is this guarantee that costs so much. If a target-benefit plan were adopted, the government's true cost of providing pension benefits could be reduced by as much as $2.5-billion a year, a savings that dwarfs the impact of eliminating income splitting.

Two questions should come to mind. First, is such a change fair? It will come as no surprise that the federal PSPP is generous relative to private-sector plans but what few people realize is that it is also significantly better than most public-sector plans. Plans such as the monolithic Ontario Teachers' Pension Plan require plan members to shoulder half the risk. If a deficit develops, the payments to fund it are borne equally by employees and employers. With the federal PSPP, the deficits are the sole responsibility of the government, which, of course, means the taxpayers.

Another way the federal PSPP distinguishes itself is its unlimited and unconditional protection of pensions against inflation. In most public-sector plans these days, inflation protection is capped. Moreover, inflation-related increases are no longer guaranteed as they depend on the funded level of the plan. As for private-sector plans, providing any inflation protection at all has become a rarity.

As a result, the average career civil servant who is retiring today after 35 years of service is walking away with a pension entitlement that is worth over $1.2-million. This might be justified if employees had been accepting lower salaries in exchange but studies show this is not the case. Cash compensation amongst the rank and file in the federal civil service is comparable with the private sector, if not on the high side. So, yes, I think most people who are not in the PSPP would agree that the change to a target-benefit plans is fair.

The second question is whether a target-benefit plan jeopardizes the retirement security of federal public-sector workers. To answer this, let's assume the current PSPP is replaced with a target-benefit plan that requires employee and employer contributions totalling 20 per cent of pay, which is the current service cost under the current federal PSPP.

A total contribution level of 20 per cent is practically unheard of in the private sector. Within a defined-contribution plan, it is not even legal, since contributions are capped under the Income Tax Act at 18 per cent of pay. Even so, a 20-per-cent plan is potentially less generous than what federal civil servants have now, but it is still very substantial. If such a plan imperils workers' retirement security, then virtually no one is safe.

If a 20-per-cent target-benefit plan earns a return that is 4.1 per cent above the inflation rate, plan members should still expect to get the same retirement benefit and inflation protection as they have now under the PSPP. Even if the long-term investment return is less, the basic amount of pension benefit is unlikely to be in jeopardy, though ancillary benefits such as inflation protection and early retirement subsidies (another plan feature that is fast becoming archaic) would be at risk. If readers think this residual risk is unacceptable, they should note that their Canada Pension Plan works in exactly the same way.

The federal PSPP is a dinosaur amongst pension plans, even in the rarified realm of public-sector plans. It is sustainable only as long as the pension promise is being underwritten by all taxpayers. The enormity of the guarantee would otherwise be a non-starter. Even politicians would grudgingly have to agree with this assessment if one is to judge by their actions in private life. To my knowledge, no cabinet minister ever came from (or went to) a private-sector company that sponsored a pension plan that is comparable to the federal PSPP.

I can see how my proposed change could be construed as an attack on public-sector employees but that would be the wrong conclusion. Outside of the federal civil service, most public-sector plans provide a more reasonable level of benefit and cost sharing. Most such plans have been gradually and responsibly trimming their promises in response to the modern-day reality of lower interest rates and lower expected returns.

As an aside, readers might wonder why the conversion of the PSPP has not already happened if it is so clearly called for. In ancient Roman times, the emperors looked to the Praetorian guards for their personal protection. To ensure their loyalty, the emperor granted the Praetorians special privileges, including a shorter period of mandatory service and higher total compensation than what the "market" required. It seems not much has changed in 2,000 years.

Frederick Vettese is the chief actuary at Morneau Shepell. His views do not necessarily reflect the views of the firm