Where covenants still have teeth: documentation in the lower middle market

Most of the 2026 conversation about credit documentation is a conversation about erosion. Cov-lite as the default, EBITDA add-backs with no cap and no time horizon, restricted payments capacity that opens up while a business is already showing stress. Spend enough time in the broadly syndicated market and you start to think that this loosening is the natural state of things.

But the lower middle market is different. In this week’s Knowledge Series webinar, I spoke with Reed Van Gorden, Head of Origination, and Natalie Garcia, Head of Underwriting, at Deerpath Capital. Both have been at the firm for close to eleven years, which makes them well placed to talk about what has changed and what has not. The headline: while the covenant architecture is the same one a broadly syndicated investor would recognize, the calibration is materially tighter.

Same vocabulary, different calibration

Deerpath is not using a different set of provisions. The building blocks are identical: a definition of EBITDA, incremental debt capacity, restricted payments, unrestricted subsidiary mechanics, the full set of LME blockers. As Natalie put it, the vocabulary and architecture are the same, but the calibration is different.

What changes is every dial. EBITDA definitions are constrained rather than elastic. Baskets carry hard dollar caps rather than ratio-based or grower formulations. The covenants are full maintenance covenants tested quarterly, not springing constructs and not incurrence-only. Unrestricted subsidiary mechanics carry far less flexibility. The terms exist in both markets - in the lower middle market they are simply locked down.

EBITDA add-backs as the cornerstone

EBITDA runs through the entire covenant package, so where you set it tells you most of what you need to know about the flexibility a borrower will have downstream. In the broadly syndicated market the add-back debate is largely over: uncapped cost savings and synergies, no time horizon, broad management discretion to define a rosy future and build it into capacity today.

Deerpath holds a different line. Add-backs have to be specific, verifiable, and non-recurring. One-time transaction costs, documented restructuring actions with a clear endpoint, identifiable cost savings with reasonable evidence. Revenue synergies are not permitted at all. Run-rate credit for new hires, sales teams, or locations a sponsor plans to open is not accepted, and where a small allowance appears it functions as a minor negotiating concession rather than a structural feature. 

Why the structure supports the discipline

The discipline is not goodwill. It rests on the structure of the market. Deerpath sits in the first lien senior secured position and is the named agent on roughly 95% of its deals, admin agent on 75% to 80%, and sole lender on about 60%. Lender-on-lender dynamics do not arise where there is no second lender to organize against. The LME blockers, the J.Crew, Chewy, and Serta protections, go into the documents as belts and braces, but the sole-lender position is what makes them rarely necessary.

That position also reflects supply and demand. By Deerpath's framing, their segment is roughly half of all deal count and close to a third of deal dollars each year, spread across a long tail of sponsors. With that much volume and relatively low competition, a lender can be selective, hold the line, and walk away from a sponsor that pushes too far. Leverage tends to run around two turns lower than up-market, and the firm targets a covenant cushion of 25% to 30% at the outset with step-downs that de-lever over time.

What stress looks like

Underwriting discipline is tested when a credit turns. The most visible signs of stress in the broader market, rising PIK toggles and amend-and-extend mechanics that defer the problem, are largely absent here. Maintenance covenants give an early trigger; monthly financials and revolver activity give an even earlier one, often before a covenant formally trips. Reed cited an estimate that cov-lite ought to carry roughly 100 basis points a year of additional premium to compensate for the value lost by not being able to step in early, a useful attempt to put a number on what covenant flexibility actually costs the lender.

When a credit does trip, a dedicated workout team takes the conversation, the originators step back, and the firm assesses whether the issue is cyclical or structural before deciding between a constructive amendment and a more aggressive posture. The going-concern logic underpins all of it: get a seat at the table early enough to force a sale, and there is usually a larger buyer willing to take on a smaller, well-run business.

The takeaway

The lower middle market is not running a different rulebook. It is running the same rulebook with the dials turned toward the lender, held in place by senior secured position, sole-lender control, and a supply-demand balance that lets a disciplined lender say no. For anyone who spends their days in broadly syndicated documents, it is a useful reminder that the architecture itself was never the problem. The calibration was.

My thanks to Reed and Natalie for a genuinely enlightening conversation.

‍