|
Gap analysis is a technique of
asset-liability
management that can be used to assess
interest rate risk or
liquidity risk.
Implementations for those two applications differ in minor ways, so people
distinguish between interest rate gaps and liquidity gaps. This article
discusses both.
Gap analysis was widely adopted by financial institutions
during the 1980s. When used to manage interest rate risk, it was used in tandem
with duration
analysis. Both techniques have their own strengths and weaknesses.
Duration is appealing because it summarizes, with a single
number, exposure to parallel shifts in the
term structure of interest
rates. It does not
address exposure to other term structure movements, such as tilts or
bends. Gap analysis is more cumbersome and less widely applicable, but it
assesses exposure to a greater variety of term structure movements.
|
|
 |
|
 |
|
The term structure of interest rates can
move in many ways. Duration analysis addresses exposure to
parallel shifts only. Gap analysis can warn of exposure to more
complex movements, including tilts and bends. |
Let's start our discussion with
cash flow matching (or simply
cash matching). This is an effective, but
largely impractical means of eliminating interest rate risk. If a
portfolio has a positive fixed cash flow at some time t, its
market value will increase or
decrease inversely with changes in the
spot interest rate
for maturity t. If the portfolio has a negative fixed cash flow at
time t, it's market value will increase or decrease in tandem with
changes in that spot rate. Stated simply, interest rate risk arises from
either positive or negative net future cash flows. The concept of cash
matching is to eliminate interest rate risk by eliminating all net future
cash flows. A portfolio is cash matched if
every
future cash inflow is balanced with an offsetting cash outflow on the same
date, and
every
future cash outflow is balanced with an offsetting cash inflow on the same
date.
The net cash flow for every date in the future is then 0.
Obviously, this is an ideal that we usually don't want to achieve, but it
is a theoretically useful concept. In its most basic form, gap analysis
assesses how
close a portfolio is to being cash matched. Here is how it
works.
Start by considering a portfolio with only fixed cash
flows—that is, the timing and amount of all cash flows is known. The portfolio contains no
floaters,
no options and no
bonds with embedded
options. Soon we will loosen the restriction against floaters, but let's
keep it for now. Gap
analysis doesn't consider credit
risk, so assume all cash flows will occur.
Gap analysis comprises aggregating cash flows into
maturity buckets and checking if cash flows in each bucket net to 0.
Different bucketing schemes might be used. As a simple example, consider a
portfolio whose cash flows all mature in less than three years. We
aggregate maturities into five buckets:
0
- 3 months
3
- 6 months
6
- 12 months
12
- 24 months
24
- 36 months
An interest rate gap is simply a positive or negative net cash flow for
one of the buckets. Exhibit 2 illustrates a gap analysis using our buckets
and some hypothetical cash flows.
|
|
 |
 |
|
Bucketed cash flows in
USD millions. A gap
is any net cash flow for a bucket, so there is a USD 100MM gap for
the 3 - 6 month bucket. There is a negative gap of USD 30MM for the
12 - 24 month bucket. |
Note that this portfolio is exposed to tilts in the term
structure of interest rates. If rates for the 3 - 6 month bucket rise and
rates for the 12 - 24 month bucket decline, the portfolio will incur a
mark-to-market loss on both gaps.
This exposure would not be identified by duration. If you calculate the
Macaulay duration of
the portfolio, it is about 0.
Now let's add floating rate instruments to the portfolio.
These generally are not bucketed according to their maturity but
according to their next reset date.
Consider a USD 100MM
floating rate note (FRN) that pays
3-month Libor flat. Its last reset was
a month ago at 2.8%. It will pay USD 0.7MM in two months, and then the
rate will be reset again.
From a market value standpoint, the FRN is equivalent to a
fixed cash payment of USD 100.7MM to be received in two months (see the
discussion of pricing in the article on
floaters). Accordingly, that is how we bucket it—we bucket the entire
FRN as a single USD 100.7MM cash flow in the 0 - 3 month bucket.
Because of how floaters are treated, buckets are often
called repricing buckets as opposed to maturity buckets—instruments are
bucketed according to their next repricing date as opposed to their
maturity date. We are moving away from cash matching and towards repricing
date matching. From this standpoint, interest rate gaps are sometimes called
repricing gaps.
So far, we have discussed the use of gap analysis for
assessing interest rate risk. It can also be used to assess liquidity
risk. Cash flows are bucketed as above. The only difference is that cash
flows from floaters are bucketed according to their maturity. The actual
values of floating rate cash flows will not be known, but estimated values
may be used. The idea of liquidity gap analysis is to anticipate periods
when a portfolio will have large cash out-flows. Such buckets are
called liquidity gaps.
A shortcoming of gap analysis—both interest rate and
liquidity gap analysis—is the fact that it does not identify mismatches
within buckets. An even more significant shortcoming is the fact that it cannot handle options
in a meaningful way. In today's markets, options proliferate. Fixed income
portfolios routinely hold caps,
floors,
swaptions,
mortgage-backed
securities, callable bonds,
etc. Options have cash flows whose magnitudes—and sometimes timing—is
highly uncertain. Those uncertain cash flows cannot be bucketed. For this
reason, gap analysis has largely fallen out of use. Today, gap analysis is
most useful as a theoretical tool for communicating issues related to
interest rate and liquidity risk.
|