Financial regulations work best when they dovetail with the underlying logic of dealers' businesses. Too often, though, well-meaning rules have unintended consequences that can undermine the economics of products and services that were never meant to be in scope.
This scenario threatened the future of equity-linked investments (ELIs), including structured notes, when the Internal Revenue Service (IRS), a department of the US Treasury, released proposed rules in 2013 clamping down on financial products that provide income contingent on US-source stock dividends.
The proposal represented a first draft of Section 871(m) of the Internal Revenue Code, which the US Treasury was mandated to write as part of the Hiring Incentives to Restore Employment Act of 2010. Its purpose: to close a loophole that allows non-US persons to receive dividend-equivalent payments from ELIs without paying a 30% withholding tax.
However, the very broad wording of the 2013 proposal would have created tremendous uncertainty around the treatment of thousands of structured notes with equity underlyings, despite the fact the vast majority of these do not offer the delta-one-like exposure or associated dividend-equivalent payments the rules were intended to capture.
When it comes to structured notes, the problem with percentage delta is you need to know what the notional of the contract is, which can be a completely arbitrary construct in a derivative
Richard Maile, JP Morgan
Enter Richard Maile, executive director in the equity exotics and hybrids business at JP Morgan in New York. Maile, a 12-year veteran of the bank, took a leading role in re-drafting Section 871(m) alongside other market participants, industry stakeholders, and US government officials. He is praised by sources close to the process for providing insights and suggestions that were enormously helpful in developing the final rules, which were published in September last year.
Maile was instrumental in developing the so-called "substantial equivalence test" included in the final draft, which assesses whether or not complex products such as structured notes truly offer delta-one-like exposure and therefore qualify for withholding tax.
"I got involved in the drafting process at the end of 2013. This was after the Treasury issued their first draft of the Section 871(m) rules where they introduced a delta test for financial products. This test stated that if the delta of an ELI was above 70%, it was taxable, and if it's less it's not. This makes sense for vanilla call and puts where delta is a good measure of how much an ELI 'looks' like a delta one, but there is an awful lot of stuff where this test does not make sense. When it comes to structured notes, the problem with percentage delta is you need to know what the notional of the contract is, which can be a completely arbitrary construct in a derivative," says Maile.
Take the example of a single-stock-linked structured note with capped two-times leveraged upside but one-to-one downside exposure. Such a note would have an indeterminate delta, as it would have 200% delta in a rising market up to the level of the cap - and 0% beyond that - while having 100% delta in a falling market. A blanket 70% delta threshold would be inappropriate in this instance, as the notional amount of stocks referenced by the note - and hence its delta - effectively fluctuates as the stock moves up and down.
So Maile went about brainstorming a solution in partnership with Treasury officials.
"We started by trying to find a way to establish to what degree a derivative looks like a delta one position without needing to know the specifics of the contract, such as its notional exposure to the underlying. This is because otherwise there could be an incentive for people to manipulate parameters, like stated notional, if in doing so they could make a contract non-taxable," says Maile.
A number of ideas were thrown around, including a test incorporating the gamma and time value of complex contracts, but these were deemed unworkable or too difficult for firms to administer. Then Maile hit upon an idea that formed the nucleus of what became known as the substantial equivalence test.
We wanted a test that could not be gamed by product issuers, that converges to a simple answer, and that didn't need an issuer to decompose a structured note into its constituent parts
Richard Maile, JP Morgan
Put simply, this test compares the hypothetical price movement of a complex ELI against the same hypothetical movement of the equities that would be used to hedge the contract - known as the initial hedge. It then measures this difference against the equivalent difference for a hypothetical simple contract with a delta of 80%, the delta threshold selected for the final rules, and its initial hedge.
If the difference in value for the complex contract does not exceed the difference in value for the hypothetical simple contract, the complex contract is deemed "substantially equivalent" to the underlying and is captured by the Section 871(m) rules.
Maile explains that this test uses information readily accessible to issuers of complex contracts and can be run using dealers' existing pricing engines, without putting undue stress on firms' IT infrastructure.
"The basic idea is you have a totally independent test which at its core tells you to what degree you expect the derivative to behave like its delta. What's attractive about this is it very much aligns the economic understanding of the trade with the tax result," he adds.
Put another way, the test rightly captures complex contracts that generate dividend-equivalent payments, but leaves out of scope those that - while referencing equities - do not truly offer investors delta-one-like exposure.
"If you want to be long 100 shares of Apple, you're not going to want to buy a digital option on Apple - because although you may save some tax as it will not be in scope of Section 871(m), you are taking on board a totally different economic exposure than if you just bought the stock outright," he says.
Maile made three trips to IRS headquarters with his colleagues in corporate tax in Washington, DC over the course of 2014 to present the substantial equivalence test to IRS officials, showing how it held up against a whole variety of scenarios and applied to a wide range of products.
"We first proposed the idea in mid-2014 and over time the IRS grew more comfortable with this test and a consensus formed around what needed to be done. We wanted a test that could not be gamed by product issuers, that converges to a simple answer, and that didn't need an issuer to decompose a structured note into its constituent parts. We are confident the final test will identify those products that should be in scope without posing an operational burden," says Maile.