[Skip to content]

Sign up for our daily newsletter
The Actuary The magazine of the Institute & Faculty of Actuaries
.

Stripping facilities

STRIPPING IS THE PROCESS of separating a couponbearing
bond into its individual cashflows,
which can then be separately held and
traded in their own right as (zero-coupon)
bonds. Official strip facilities have been available in
the United States for some Treasury securities since
1985, and official strips markets have also been established
in other countries such as Germany and Spain.
Although anyone can trade or hold strips, only a giltedged
market-maker (GEMM), UK Debt Management
Office (DMO), or the Bank of England can strip a strippable
gilt. As with conventional gilts, GEMMs are
obliged to make a market in strips. Gilt strips are registered
securities and, as an obligation of the UK government,
have the same credit risk as ordinary gilts.
By the end of December 1999 the number of strippable
bonds was 11, totalling £116bn and representing
over one-third of the total amount of gilts
outstanding. The 11 strippable issues have aligned
coupon payment dates on 7 June and 7 December.
This means that investors in gilt strips have a choice
of two dates each year on which to receive cashflows.
Coupons from different strippable bonds that are paid
on the same day are interchangeable (or ‘fungible’)
when traded as strips, thereby enhancing the liquidity
of the market. However, in response to market
feedback, the Bank of England decided not to permit
fungibility of coupon and principal strips. Figure 1
below illustrates the maximum potential size of each
coupon and principal strip in December 1999.
Despite the large pool of bonds eligible for stripping,
the market in gilt strips has grown slowly. By
the end of 1999 just £2.8bn (or 2.4%) of strippable
gilts were held in stripped form and weekly turnover
in gilt strips averaged around £70m, compared with
around £15,305m in non-strips. Factors that have
contributed to this slow development have been the
need for pension fund trustees to give the appropriate
authority to fund managers to invest in strips,
and the continuing inversion of the yield curve over
most of 199899 making strips appear more expensive
relative to conventional gilts. The turbulence in
financial markets during 1998 also discouraged strips
activity as investors sought the liquidity of the underlying
gilt market. Retail demand for strips has been
hampered by the necessary tax treatment, which stipulates
that the securities are taxed each year on their
capital gain or loss, even though no income payment
has been made.
Features and uses of strips
As zero-coupon instruments, strips can be considered
financial building blocks which can be used to create
a variety of synthetic assets whose cashflows cannot
be produced with a combination of conventional
gilts, eg annuities or risk-free savings products for definite
future needs (eg tuition fees).
Strips may be attractive to investors because of the
lack of reinvestment risk, making them useful in helping
to immunise a portfolio against interest rate risk.
Other important attributes are that they typically
have higher duration than coupon-bearing bonds and
are also much more convex than similar maturity
coupon bonds. These features mean that they allow
active fund managers to take greater exposure to price
movements and that strips should outperform
coupon-bearing bonds of the same maturity, both
when prices rise and when they fall.
Assessing relative value between gilts
and gilt strips
The introduction of the gilt strips market means that
it is now possible to observe traded zero-coupon rates
directly in the market rather than having to obtain
theoretical zero-coupon rates from a yield curve
model applied to conventional bond data. Figure 2
compares the theoretical zero-coupon curve at end
1999 with the corresponding zero-coupon curve
obtained from coupon strip yields, plus the par yield
curve from conventional bonds, as well as yields on
principal strips. Principal strips typically trade at a
price premium (lower yield) to the corresponding
coupon strips because they are necessary for reconstitution
and are generally larger so more liquid.
In the situation where the yield curve is upward
sloping, the strips curve will always lie above the par
yield curve for coupon-bearing gilts, as the value of a
zero-coupon gilt is only determined by the discount
rate applicable to the maturity payment, while the coupon-bearing gilt valuation is affected by
the regular coupon payments, each of
which is discounted at a lower rate than the
maturity payment (given the upward sloping
curve). Conversely, in a downward sloping
yield curve environment, the strips
curve will always lie below the par yield
curve. Superficially, this makes strips look
less attractive than coupon gilts. However,
when assessing whether strips genuinely
provide value relative to gilts, it is not
enough to simply compare the yield on a
strip with that on a strippable coupon gilt
of the same maturity. This takes no account
of the differences in duration between the
two instruments.
Comparing the yields on coupon gilts
with the yields on strips of the same duration
provides a much better indication of
relative value between the sectors. In a downward
sloping yield curve environment, since the yield on a
strip will be higher than that on a bond of the same
duration, one trading strategy would be to sell the
bond and buy the strip. Table 1 illustrates the yield
pick-up achievable by executing a trade of this form
using data from end December 1999. In each case, the
strip purchased is that of closest duration to the bond
sold. Clearly, by using a combination of strips rather
than a single strip, it would be possible to get a precise
duration match.
Another measure of relative value between bonds
and strips is obtained by comparing the value of a
coupon gilt with the combined value of all the strips
that make up that bond. This will indicate whether
the arbitrage condition between bonds and strips
holds (ie that the price of the bond should be the sum
of the prices of the bond’s components) and hence
will highlight if there are any pricing discrepancies
between the sectors and, in particular, whether there
is an incentive to strip or reconstitute the gilt. For
example, data for end 1999 suggest that the price of
6% 2028 was about 60p lower than the total of the
component strip prices (equivalent to about three
basis points in yield terms). Ordinarily, you would
expect this difference to be much smaller, suggesting
that the size of the spread here probably reflects a
reduction in liquidity and quality of price data over
the Christmas vacation period rather than a genuine
arbitrage opportunity.
Future developments
At present, all strippable gilts are fixed-coupon conventional
bonds. One possible development of the
gilt strips market that the DMO keeps under review
would be to permit index-linked gilts to be stripped.
With no reinvestment risk, index-linked strips would
provide investors with an instrument that offers a
guaranteed real return and so should be attractive to
investors that desire real value certainty. Although
many government bond markets have the market
infrastructure in place that enables participants to
strip conventional bonds, only Canada, New Zealand,
and the US currently permit the stripping of indexed
bonds. Despite the potential attractions of indexlinked
strips there has been little or no stripping in
these markets. In the UK, index-linked gilts are not
currently seen as potentially attractive for stripping
purposes because of the diverse coupon dates. This
would limit coupon strip fungibility, and would be
likely to result in illiquid securities.
Liquidity in the conventional strip market is aided
by the DMO’s current policy of making all new issues
of conventional gilts strippable. Additional measures
introduced recently to bolster the size of benchmark
gilts should also help to build up liquidity in strippable
gilts and hence the strips market. These
include the reintroduction of conversions in 1998
and the launch of the first switch auction in October
1999. Looking forward, with the likely reduction in
the need to raise revenue through gilt sales, switch
auctions are likely to become increasingly important
in maintaining liquidity in the market. Another
development that the DMO has discussed with market
practitioners, and which it keeps under review, is
the introduction of a second set of strip cashflow
dates. This would increase investor choice regarding
when they wish to receive cashflows but could
detract somewhat from liquidity in coupons, since it
would almost certainly lead to a reduction in the
quantity of bonds issued with the original strip cashflow
dates.

00_08_05.pdf