This website uses cookies, pixel tags, and similar technologies (“Cookies”) for the purpose of enabling site operations and for performance, personalisation, and marketing purposes. We use our own Cookies and some from third parties. Only essential Cookies are used by default. By clicking “Accept All” you consent to the use of non-essential Cookies (i.e., functional, analytics, and marketing Cookies) and the related processing of personal data. You can manage your consent preferences by clicking Manage Preferences. You may withdraw a consent at any time by using the link “Cookie Preferences” in the footer of our website.
Our Privacy Notice is accessible here. To learn more about the use of Cookies on our website, please view our Cookie Notice.
The grammatical quirk that impacts cost savings in debt covenants
James Wallick
+Antony Serban
•5 min read
The nature and scope of what companies add back to EBITDA has broadened significantly over the past few years, to the point that modern-day debt covenants will often reveal three or more pages of expected cost savings and other items that can be added back to boost earnings.
Sometimes it feels like there’s no limit to what companies and their lawyers can come up with. But when it comes to the timeframes for achieving some of these cost savings, good old-fashioned English grammar might impose a hard stop.
Note the difference between these two sentences:
We expect to realize cost savings within 24 months
We expect to realize cost savings resulting from actions we expect to take within 24 months
The first sentence clearly contains a timeframe for expected realization of the savings. In the second, however, the company should argue that the 24-month timeframe refers only to the actions(for example, drafting a cost saving plan) as opposed to the actual realization.
For an extreme example, let’s imagine a company expects to implement a cost savings plan within the next 24 months, but that it would take four years to actually realizethe savings. That’s a six-year lead time for the cost savings — but arguably, it’s permissible by the formulation of the second sentence.
When buysiders find themselves in a position to negotiate covenants, shortening these cost savings timeframes is often quite high on their list of demands (for more on this, see our Covenant Pushback tool for European high yield bonds here).
However, the linguistic details of how this shortening is represented in the final docs seem to receive less attention. It seems to us that investors are often satisfied if they see a time horizon cap of some sort, but sometimes overlook the underlying language.
Is time on your side? It depends on the grammar, and what the courts make of it.
The “Last Antecedent Rule”
Since as early as 1799, courts in the United States have recognized the “last antecedent rule.”
The US Supreme Court has described the clause as follows: “a limiting clause or phrase … should ordinarily be read as modifying only the noun or phrase that it immediately follows…. [Further, while] this rule is not an absolute and can assuredly be overcome by other indicia of meaning, we have said that construing a statute in accord with the rule is ‘quite sensible as a matter of grammar.’” [internal citations omitted]. Barnhart v. Thomas, 540 U. S. 20, 26 (2003).
The New York Court of Appeals has a similar description: “the last antecedent rule of statutory construction applies here. Under that rule, ‘[r]elative and qualifying words or clauses in a statute are to be applied to the words or phrases immediately preceding, and are not to be construed as extending to others more remote’” [internal citation omitted]. Matter of T-Mobile Northeast, LLC v DeBellis, 32 NY3d 594, 608 (2018).
In Barnhart, Justice Scalia illustrates the rule with the following plain English example: “Consider, for example, the case of parents who, before leaving their teenage son alone in the house for the weekend, warn him, ‘You will be punished if you throw a party or engage in any other activity that damages the house.’ If the son nevertheless throws a party and is caught, he should hardly be able to avoid punishment by arguing that the house was not damaged.”
In other words, the limiting clause “that damages the house” applies only to the immediately preceding phrase “engage in any other activity” and not the more remote phrase “throw a party.”
Let’s revisit the cost savings formulations we highlighted earlier:
Example A:
Example B:
In Example A, it is clear that the timeframe concerns realization. But in Example B, under the rule of the last antecedent, the words “within 24 months” would be interpreted to apply to “actions either taken or expected to be taken,” rather than “result”.
Data from H1 24 High Yield Bonds
Many deals we see have a timeframe for action. But in our H1 24 High Yield Bond Trends report (coming soon!) we split the 52 deals — including 22 sponsor-backed deals — in our dataset into two categories, bearing in mind the last antecedent rule and its impact on action versus realization:
Total deals
Timeframe for Realization: 12%
No timeframe for Realization: 88%
Sponsor-backed deals
Timeframe for Realization: 5%
No timeframe for Realization: 95%
For deals with a timeframe for realization, the period ranged from 12 to 24 months, with the average time frame being 15 months. We admit this is a small sample size.
You can’t always get what you want
Today’s debt markets continue to see boundary-pushing EBITDA add-backs, which often afford companies more and more capacity to raise additional debt and pay dividends.
If the market wants time caps on cost savings add-backs to apply to savings that are actually realized (as opposed to just being planned or worked towards) within a given time period, a closer focus on drafting could go a long way.
Investors may not be able to rein in inflated EBITDA completely. But if you try sometimes, you might just get closer to what you need.
