Less Than Meets the Eye: Equity Linked CDs

It sounded too good to be true, which should have tipped me off: A five-year equity-linked CD that offers a return-of-principal guarantee backed by the U.S. government, coupled with 100% of the average appreciation of the S&P 500 index during that time period. What a deal! A downside guarantee and all that upside!

A closer look, however, reveals somewhat less sizzle than the hype would suggest.

Equity-linked CDs are issued by a bank, so return of principal is guaranteed by the FDIC. But the term "100% of the average appreciation" is more than just a tad misleading. I'd be surprised if everyone who buys the CDs understands exactly what they are getting. Let's say you invest in such an instrument when the S&P index is valued at 1000. Five years later, the value is 2000. You might expect that your equity-linked CD has doubled in value. [(2000-1000)/1000=100%]

Not so. To simplify only slightly, the S&P 500 index's value is fixed on the date of the CD's issue. For the next five years, the index is measured on the anniversary of that date. Then, those index numbers are averaged to determine your return.

In our example, the initial value is 1000, and let's say on the date of issue over the next five years the index reads 1200, 1400, 1600, 1800, and 2000. The average of those five index values is 1600. That average figure is used as the end value to determine your profit [(1600-1000)/1000=60%]. Your profit is thus 60% (see the first example in the table below).

That's a far cry from the return you might have anticipated. The phrasing "100% of the average appreciation" is valid from a mathematical perspective, and perfectly legal to put in the prospectus. It's also a marketer's dream.

(Incidentally, the numbers above are for simplicity only. I am not suggesting that I think the S&P 500 is liable to double in the next five years. That would be considerably above its long-term performance, and after its recent staggering run up in recent years, a further doubling is unlikely.)

Hypothetical Performance of Equity-linked CDs
year 1 value	year 2 value	year 3 value	year 4 value	year 5 value	average value	% return
1200	1400	1600	1800	2000	1600	60
2000	2000	2000	2000	2000	2000	100
500	500	500	2000	2000	1100	10
All examples assume initial S&P value of 1000.

Of course, the path that the S&P takes while getting from 1000 to 2000 will affect the outcome. If it were to shoot right to 2000 the first year and then stay there for the remaining four, the results would look a lot better than if it fell 500 points the first year, stayed there for two more years, and then shot to 2000 (see examples above).

What about dividends? Forget it. You don't get them. Dividends are not part of the index's value.

But at least we've got downside protection. Surely that no-loss guarantee is worth something, right? Well, maybe. If we look at all 69 five-year periods since 1926, we see only seven periods of losses (five during the depression). That's only about 10% of the time. (Of course, those losses can be significant--they ranged from 48.63% to 1.01%, with the average being 28.84%.)

It's not hard to figure out what the bank is doing to put this deal together. Like most other banking operations, the bank isn't putting itself at risk. First, they buy enough zero-coupon bonds to meet the principal guarantee. Then, after everybody gets their cut of the fees, the balance goes into long-dated call options on the S&P index to provide for the appreciation potential. When interest rates are low, most of the initial deposit goes into the zero-coupon bonds. When market volatility is high--like now--these call options are expensive. So, with options expensive, and little to spend on them, we can't get 100% of the upside.

I'm a full-disclosure kind of guy, so I looked through the prospectus of one of these to see what the fees were. Unfortunately, this being an initial offering, the fees have not been disclosed. Fees do matter because they come out first, before the options are purchased. More fees mean fewer option contracts and less upside potential for the investor.

I did find out that during the same time period a competing broker was offering a similar deal with 105% of the average appreciation. From that we can surmise that this particular offering was a pretty good deal for the brokerage, and that it pays to compare.

One risk to consider: The investor is giving up the risk-free return--say 5%--he could have earned in the CD, T-Bill, or zero-coupon fixed-income security, which would be a sacrifice if the S&P 500 doesn't appreciate sufficiently. This opportunity cost must be weighed against the limited upside potential. In other words, for the investor to "win" with an equity-linked CD, he must obtain at least a 5% return per year. That may not seem hard to achieve when the market is rallying, but the averaging provision and the dividend exclusion might be formidable barriers in slower markets.

Whatever you think of the deal in itself, it's not particularly attractive for taxable accounts. Somebody at the brokerage or bank guesses what the annual return will be, and the investor must pay taxes on this theoretical return each year even though he has no cash in his hand. At the end of the five years, the investor and Uncle Sam settle up for the difference. Ordinary income tax rates are used, so the investor gives up both deferral and capital gains treatment on his returns.

Are equity linked CDs a good deal? Perhaps for some investors. But, reviewing this offering reminded me of those words of wisdom that my mother used to whisper as she bounced me on her knee: "Son, there ain't no such thing as a free lunch." So true, Mom!