with the effectiveness of most LDI strategies as plans achieve higher
funding ratios and seek precision in their asset/liability match. The
primary reason is that the benchmarks used in traditional de-risking
strategies, namely the Barclays Long Credit Index and the Barclays
Long Government/Credit Index, simply were not created for pension
plans. They are subsets of the Barclays Aggregate index, which
represents bonds outstanding, which in turn means these indexes
really represent issuance patterns, not a pension plan’s liability. In the
case of the Barclays Long Credit Index, for example, over 49% of the
index is made up of BBB-rated securities. In the case of the Barclays
Long Government/Credit Index, it’s about 29%. The latter index also
has a substantial amount of Treasury securities. By contrast, pension
liabilities are generally discounted, for funding purposes, using an
index of A-rated to AAA-rated bonds and a discount curve derived
from them. The upshot is that traditional market-based benchmarks
have a lot of tracking error against pension discount curves—around
6% according to our research.
CIO: How has asset/liability tracking error impacted pension
Wilson: It’s been a significant issue, especially given that funded
ratios remain so low for most corporate plans. Because of low
funded ratios, many plan sponsors hold a certain amount of
return-seeking assets to gain ground on their unfunded liability.
As a result, even after all the talk and effort put into de-risking,
the average asset allocation for a corporate pension plan today is
still close to 60/40, meaning 60% return-seeking assets and 40%
fixed income or liability hedging assets. The risk a 60/40 portfolio
presents on an asset/liability basis is quite substantial, especially
during adverse market conditions. For example, in 2006, when
markets were booming and volatility was low, a 60/40 portfolio
had approximately a 5% annualized tracking error against a typical
pension liability, which isn’t so bad. But in 2008, which was obviously an extraordinarily difficult year, that tracking error increased
to 45% annualized. The last thing you want to happen while you’re
pumping money into your pension plan is to have this tracking error
further derail your de-risking strategy. Merely extending the duration of your fixed-income allocation to better match a portion of
your liability’s duration, which is all that many plan sponsors do,
doesn’t really have a significant impact on overall plan risk. You also
have to address the return-seeking side of the portfolio.
CIO: What tools can corporate plans use to manage risk?
Wilson: There are a wide variety of tools available to plan spon-
CIO: Once a plan sponsor is ready to fully de-risk, how can
sors. One we like, for plans that have large equity allocations, is a
managed volatility approach in which we identify the exact amount
of risk, from a volatility perspective, that the plan is willing to take,
and then manage to that target using a futures-based overlay. As
an example, let’s say the plan has exposure to a large-cap equity
strategy and needs to maintain a growth portfolio because the plan
is underfunded—but can’t withstand the substantial volatility risk
that the asset class presents. We might set an annualized volatility
target of 12% for that growth portfolio, and create a volatility band
around that from, say, 10% to 14%. When we forecast volatility to
be above our 14% threshold, we’ll sell equity futures and bring the
market exposure down and volatility back within the range.
they minimize tracking error? Can't derivatives help?
Wilson: Derivatives are anathema to many plan sponsors, especially
smaller ones. But there is a simpler way. At Nuveen, we worked in
collaboration with Wilshire Analytics to develop a family of four
indexes, introduced just last year, designed specifically for corporate
pension plans. The Nuveen Wilshire Pension Investment Indexes seek
to provide an effective de-risking solution, customized for all plan
sizes, with a simple implementation. They also serve as both an asset
benchmark and a liability benchmark. They enable a plan sponsor to
judge whether the bond manager is adding value against the benchmark and whether the LDI program is effective.
CIO: How do these new indexes work, and how do they
contrast with traditional pension plan benchmarks?
Wilson: The indexes cover the corporate universe of A-rated bonds
and better, which means they closely match the credit quality of
pension discount curves. They are distinguished from one another by
their maturity range (e.g. 5 to 10 years, 10 to 20 years, and 20 to 30
years), because if you think about a pension liability, it’s a series of
cash flows that can be bucketed very similarly to the way we bucketed
our indexes. We also developed an ultra-long STRIPS index solely for
extending duration when needed against a plan’s long-dated liabilities—without needing derivatives. Also, one of the things we were
trying to solve for during the development process was allowing
smaller plans to have access to the same level of customization that’s
long been available to their larger peers, using products they know
and can understand. So, earlier this year, we launched a series of
collective investment trusts managed against each of these indexes,
with a $1 million minimum account size. This means even very small
corporate pension plans can get the same level of customization and
precision available to much larger plans.
CIO: Let’s turn to public pension plans. How are their chal-
lenges different from those facing corporate plans?
Wilson: Public plans and corporate plans are fundamentally the
same at their core. It’s the differences in their accounting and regulatory treatment that make them different from an investment strategy
perspective. Corporate plans value their liabilities using a market-based yield curve, which means their liabilities carry duration and
credit spread premium risks that need to be managed. Public plans,
generally speaking, have a more static, non-market-based discount
rate—which currently averages about 7.5%—which means their
liability does not have any duration or credit spread risk.
Beyond these distinctions, public plans are even more underfunded,
and that presents some unique challenges for them. To be fair, many