Morningstar - Q3 2020 - 41

multiple scenarios with the goal of generating
the probabilities of different outcomes; this
type of analysis is often beyond the reach
of individual investors or brokers but is bread and
butter for sophisticated banks. In our analysis,
we used 20 years of historical daily price
index returns to calculate 10,000 different
scenarios and then determine the probability
of a given outcome.
Even so, we had to make some simplifying
assumptions. As noted above, many structured
notes-including the Goldman Sachs issue
we used as our test case-can be called before
maturity. It's impossible to predict if or when
this might happen, so the note could have
a maturity as short as three months or as long as
two and a half years. So we had to simplify
our analysis by assuming the note will last until
its stated maturity, which reduces the number
of potential payoff paths.
As a reminder, this note has a $1,000 par value
and promises to pay $22.75 per quarter depending
on the performance of its underlying indexes.
Assuming the note lasts until maturity, there are
10 potential payments for this note-one for
each quarter before the maturity date plus the
payout at maturity. Each quarterly coupon payment
may or may not happen, depending on whether
each index stays above the coupon threshold
or not. The amount of principal paid back at
maturity also varies depending on the return of
the worst-performing index. With different
possibilities for the final principal payoff (either par
or some value less than par), we end up with
2,048 payout possibilities.2
Using historical daily price returns of the three
indexes in question to estimate their means,
volatilities, and covariances,3 we ran a Monte Carlo
simulation to estimate all the potential payouts
from this note.4 We then summed up the payouts
and looked at the frequency of how often each
total appeared as a result. E X H I BI T 1 shows the

distribution of the results from the 10,000 different
scenarios in our simulation.
What we see here is that slightly less than half
of the paths result in full payment of 10
coupons and the return of principal. The majority
of the paths result in a smaller, sometimes
significantly smaller, sum of payments. The
distribution has a long left tail, with non-negligible
probability of suffering losses of 25% or more.
Because of that, the expected value of all the
payouts stands at $1,042, which translates
(roughly) into a 1.67% mean rate of annualized
return. That's quite a difference from the maximum
9.1% rate of annualized return the investor is
teased into expecting by the note's coupon.
Fees and Expenses
The costs of structured products are both relatively
high and difficult to understand. In contrast to
standard investment products such as mutual
funds, which come with clearly disclosed annual
fees or up-front commissions, the true costs of
structured products are typically hidden.

Structured products include several types of
fees, both explicit and implicit. The explicit fees are
reported as underwriting fees and discounts.
In our sample group of about 50 recently issued
products, we found that reported fees averaged
about 2.0% and ranged from zero to as high
as 4.1%. These fees are generally paid by the issuer
to the selling broker as an up-front fee. (While
investors don't pay underwriting fees and
discounts directly, they're built into the price
of the issue as a markup and are ultimately paid
by the end investor. Investors purchasing
notes for a fee-based account can sometimes pay
a lower purchase price to reflect the lack of
commission payments.)
While the mechanics of these fees are
straightforward, the true cost of structured
products is much higher. Par value is one
confusing aspect of these products. Issuers report

this as either a principal amount or par value,
which makes it sound like investors are getting
a traditional fixed-income product, such as
a bond or a certificate of deposit. But unlike the
face value of a bond, the par or principal
value of a structured product doesn't really reflect
the value of the investment investors are
purchasing; it's just the minimum amount that
can be purchased.
In fact, the actual value of the product will almost
always be lower than the stated principal value.
A structured product is typically made up of
other securities, such as a zero-coupon bond plus
some combination of put or call options. These
components can be valued by using standard
financial tools, such as present-value and
option-pricing formulas. The difference between
the underlying value of a structured note and the
principal/par value is basically a markup or
embedded fee.
In 2013, the SEC began requiring issuers to report
estimated fair values as part of their registration
statements. These estimated fair values will
typically be reported in the term sheet, or the final
version of the registration statement that's
filed when the notes are issued. They're shown as
a dollar value, which is based on the principal
value of the note. For example, an issuer might
report an estimated value of $975 for a note
sold in increments of $1,000, or $9.80 for a note
sold in increments of $10.
In our sample set of recent issues, we found that
reported fair values averaged about 97.1% of
the par value. In other words, there's an average
markup or embedded fee of 2.9%.
Call Risk and Lack of Symmetry
Call risk is another issue investors should be
aware of. A large percentage of structured note
issues are callable, which means issuers
can redeem them before the maturity date. Nearly
three fourths of the recently issued structured

2 2^11.
3 We assumed a multivariate log-normal distribution for the purpose of the simulation.
4 Alert readers will note we are not assuming the reinvestment of the coupons into the underlying instrument, which is the standard way of calculating total returns. That's because one can't
reinvest the coupons into the note itself, which is sold in $1,000 par increments and is illiquid anyway.

morningstar.com/lp/magazine

http://www.morningstar.com/lp/magazine

Morningstar - Q3 2020

Table of Contents for the Digital Edition of Morningstar - Q3 2020