Many pension plans operate on the assumption that they will earn 7.5% to 8.5%
on their combined stock and bond portfolios, after expenses. These assumptions
seem too optimistic.
Some companies have been reporting unrealistically low pension expenses. For
example, in my opinion, it defies common sense to see that CN has reported a pension expense of
just $22 million in 2004 while actually contributing $165 million. These pension
expenses seem sure to rise materially.
Companies will have to increase their pension contributions and pension
expenses to eliminate the existing pension deficits even if they do in fact earn the
7.5% to 8.5% going forward. And if they lower those return assumptions then they
will automatically face even larger pension expenses.
The result will be higher pension contributions for employees and companies
and lower profits for companies. In the worst cases the pension plans will suck
every dollar of profit out of some companies and even that may not be enough.
Some companies will go bankrupt over this issue and the pension plans will be
turned over to trustees who will be forced to reduce pension benefits to
retirees. A vicious circle may result where newer companies with no pension
deficits will out-compete the old-line companies and hasten their demise.
Another result is that many companies will stop offering defined benefit pension
plans to new employees. They are just too risky for the companies.
As if the problems of lower stock returns was not bad enough, low interest
rates may be an even bigger problem.
rates already have and are still dramatically increasing the value of the pension liabilities. The result is that
many pension deficits will continue to soar. This has already been happening but
the impact in 2005 may be larger than in recent years (although the strong
Canadian stock market performance in 2005 will offset this problem to some
Long term interest rates were still dropping like a stone during much of 2005
- even as short term interest rates were rising. The 10 year
government bond rate in the United States was recently 4.53% and in Canada was 4.10%. High
quality corporate bond yields were at about 4.70% to 5.0% for 10 year issues,
and this probably represents a reasonable weighted average across the spectrum
from short to very long bonds.
While long-term interest rates used in the 2004 annual reports were low,
Canadian long-term interest rates have moved materially lower as of December
2005. These lower long term interest rates will lead to even higher pension plan
deficits. A pension deficit exists if the value of plan liabilities (pensions)
discounted at the appropriate rate is lower than the value of plan assets.
In Canada, the Canadian Institute of Chartered Accountants handbook at section
3461 requires that the discount rate be based on high quality corporate bond
yields with maturities matched to the pension obligations. And guess what?
today's dramatically lower high grade corporate bond yields used to discount the future
pension liabilities automatically leads to a dramatically higher value of those
liabilities. When those higher liabilities are subtracted from the same level of
plan assets, voila, you get a huge increase in the deficit. Or you move from
surplus to deficit. To demonstrate, imagine that the weighted average liability
amounts to $100 million to be paid in 15 years. Discounted at 7% this represents
a present value of $36.2 million. But discounted at a lower interest rate of 5.0%, the liability rises to a present value of $48.1 million. This represents a
huge 33% increase in the value of the liabilities.
Now, it appears that many Canadian companies have not yet adequately reduced their
discount rates for pension liabilities. For example most of the discount rates
noted in the table above are still well above the 4.7 to 5.0% level of current
high-quality 10-year corporate bond yields. I expect
that under the handbook, corporations have some leeway to use expected corporate bond
yields rather than the spot rates at their year-end.
In fairness, using corporate bond yields as the discount rate is
conservative. Normally, one might use the (higher) expected return on plan
assets as the discount rate. But the Canadian law states that high quality
corporate bond yields must be used. So it appears that if interest rates remain
at recent very low levels or decline further, companies will be forced to lower
their discount rates thereby increasing their pension liabilities.
From the example above, you can see why companies would be reluctant to lower
the discount rate. Since the end of 2004, interest rates have dropped a further
significant amount. In reporting their 2005 results many companies may be forced
to cut their pension discount rates substantially.
So, in spite of strong stock market returns in 2005, pension deficits will probably soar
with the 2005 annual reports because of
the dramatic drop in interest rates and (more specifically) in high grade
corporate bond yields.
And there is more bad news. The declining interest rates have actually led to
large capital gains in the bond investments held by pension plans. These capital
gains have been adding to the pension assets and have hidden the true extent of
the unfolding disaster. At some point interest rates must bottom out. At that
point the capital gains must abruptly stop. At that point the pension liability amounts will
be maximized by the low discount interest rate (high-grade corporate bond yield) used in the calculation. And at
that point yields on high grade corporate bonds will be near historic
lows of under about 5% for 10 year bonds.
Now a typical pension plan may have 40% of its funds in bonds. At that point
with stocks returning perhaps 7 to 8% and high-grade corporate bonds returning
perhaps 5% on average, pensions will
have a tough time forecasting more than a 6% - 7% overall average return. This
will automatically lead to higher pension expenses and reduced profits.
Many defined benefit pension plans are probably headed for an absolute disaster. When
the extent of the damage that will be and has been caused by extraordinarily low
long-term interest rates becomes known, the stocks of some of these companies
could be hit hard. As an investor I will combat this by generally avoiding
investing in large companies that have large defined benefit plans. I will
particularly avoid those companies with a high ratio of retired workers compared
to active workers.
A possible solution?
Some observers blame pension deficits on the inherent uncertainty of
predicting stock market returns. (These people may not have noticed that in
reality today's incredibly low interest rates are the prime contributor to
pension deficits). One solution proposed is for pensions to invest strictly in
bonds and avoid stock investments. Incredibly then this solution proposes to
solve the deficit problem which was caused to a large degree by low interest
rates on bonds by restricting investments to only bonds!
In the long-term corporations are wise to get out of the business of
guaranteeing pensions and other post-retirement benefits of workers. Such
guarantees amount signing a blank cheque. These defined benefit plans should
(if possible) be
essentially contracted out to large insurance companies who are in the business
of managing such risks. Otherwise, if corporations cannot contract out their
defined benefit pension plans then I believe that they would be wise to convert
to defined contribution type pension plans.
If there is any hope of funding pensions at a reasonable cost, then pensions
will have to continue to invest a significant portion of their assets in stocks.
After all, stocks are expected to return more than bonds.
For existing pension plans the expected returns on plan assets as well as the
discount rate assumptions need to be realistic. The accounting rules may need to
be changed. It seems overly conservative to calculate pension deficits on the
assumption that all plan assets would be invested in bonds, this accounting
requirement may need to be challenged. For most existing defined benefit
pension plans, the contributions of both employees and the employer is likely
going to have to rise.
Shawn Allen, P.Eng., MBA, CMA, CFA
June 15, 2003 (Last revised December 10, 2005)