# Debt Facilities, Covenants, and Bank Relationships
In March 2023, Silicon Valley Bank failed in 48 hours. Less discussed is what happened inside its corporate borrowers over the following week. CFOs who held undrawn revolvers with SVB discovered a brutal truth: a committed facility is only as good as the counterparty's ability to fund it. Several drew down their entire lines pre-emptively—not because they needed cash, but because they understood that liquidity you have not yet claimed is liquidity someone else controls. The CFOs who slept well that week were the ones who had spent the prior two years diversifying their banking group, negotiating their covenant packages loose, and building relationships deep enough that they got a phone call before the market did.
That is the real subject of this lesson. Not what debt facilities *are*—you know that—but how they are *engineered*, why the covenant grid is where control actually lives, and how a CFO orchestrates a banking syndicate so that credit is a standing asset rather than an emergency scramble.
Most CFOs inherit a capital structure and treat the debt documents as fixed infrastructure. The sophisticated ones treat every facility as an instrument they are actively engineering across four dimensions: tenor, tranching, flexibility, and optionality.
Start with tranching. A single term loan is a blunt instrument. A well-built structure segmentssegmentsDividing a market into distinct groups of customers who share similar needs, characteristics or behaviours, so each group can be served with a tailored approach. debt by purpose and by the lender base that prices it best. A revolving credit facility (RCF) sits at the top for working-capital swing and liquidity backstop—priced tight, rarely drawn, and provided by your relationship banks who want the ancillary business. Beneath it, a Term Loan A (amortizing, bank-held, tighter covenants) funds a known capital need. A Term Loan B (bullet, institutionally held, covenant-lite) funds larger or longer-dated needs where you want minimal maintenance testing. The judgement call is *matching the instrument to both the asset and the investor appetite*—you do not fund a 7-year program with a 364-day facility, and you do not put a maintenance covenant grid on capital that institutional investors will happily hold covenant-lite.
The distinction that separates amateurs from operators is understanding committed versus uncommitted, and drawn versus available. An uncommitted line is a courtesy the bank can withdraw. A committed facility, with a commitment fee paid on the undrawn portion, is a contractual obligation to fund—subject to no default and to the "no material adverse change" (MAC) representation being true at drawdown. That MAC clause is the trapdoor. Read it. In a genuine crisis, a lender's counsel will scour the MAC condition to avoid funding. The best CFOs negotiate the MAC narrowly at signing—tied to specific, objective triggers rather than a vague "material adverse effect on the business"—precisely so the facility funds when they most need it.
Then there is accordion capacity (also called an incremental facility): pre-negotiated headroom to upsize the facility later without reopening the full syndication. If you expect to grow, you build the accordion in now, when you have leverage and the credit is clean. Negotiating incremental capacity mid-crisis is negotiating from weakness.
The practical Monday-morning discipline: maintain a facility map—a one-page view of every tranche, its tenor, its maturity, its pricing grid, its undrawn availability, and its provider. If you cannot draw this from memory, you are not managing your debt; it is managing you.
Here is the framing that reorganizes how you think about debt: the coupon is the price of the money, but the covenants are the price of your freedom. A CFO who negotiates 25 basis points off the margin while accepting a tight leverage covenant has traded a rounding error for the right to run the company.
Covenants fall into three functional categories, and you manage each differently.
Maintenance covenants are tested every quarter regardless of what you do—typically Net Debt/EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition → below a ceiling, or Interest Coverage (EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition →/Interest) above a floor. They are the tripwires. Incurrence covenants are only tested when you *take an action*—raising more debt, paying a dividend, making an acquisition. Bank facilities (RCF, TLA) carry maintenance covenants; institutional debt (TLB, high-yield bonds) leans on incurrence covenants only.
The strategic implication is direct: the more maintenance covenants you carry, the more often your lenders get a vote on your future. Every quarter you are re-earning the right to your own capital structure. A covenant-lite structure defers that conversation until you choose to act. This is why, in benign credit markets, strong issuers push hard for incurrence-only packages—they are buying freedom, not just cheaper debt.
Amateurs negotiate the covenant *level* (leverage of 3.5x vs. 4.0x). Professionals negotiate the definition of EBITDA, because the definition determines how the level is actually calculated. This is where the real game is played.
Consider "Adjusted EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition →." Does your definition permit add-backs for run-rate cost synergies from an acquisition (giving you credit today for savings you have not yet realized)? Does it cap those add-backs at, say, 25% of unadjusted EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition →, or leave them uncapped? Does it permit pro-forma treatment of a full year of an acquired business's earnings? Each of these can move your leverage ratio by half a turn or more. A 4.0x covenant against a generously defined EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition → can be looser than a 4.5x covenant against a punitively narrow one.
The Monday-morning skill is to build your covenant compliance model the way your lenders will build theirs, then stress it. Run your base case, your downside case, and your severe downside. Identify the quarter and the scenario in which you breach. That date—your covenant runway—is one of the three or four numbers you should be able to recite instantly.
You do not run at your covenant limit; you run against a self-imposed cushion. If your leverage covenant is 4.0x, you manage the business to stay at or below 3.25x, giving yourself roughly 20% headroom. That cushion is your margin for a bad quarter, an inventory build, or a demand shock.
When the cushion erodes, you have a hierarchy of responses—and the discipline is to act *early*, while you still have options:
1. Operational levers — accelerate collections, defer discretionary capexcapexCapital Expenditure (CapEx) is money spent to acquire, upgrade, or extend long-lived assets like equipment, property, or software that deliver value over multiple years.View full definition →, unwind a working-capital build. These fix the ratio without involving lenders.
2. Structural levers — an equity cure (many facilities permit a cash equity injection to be counted as EBITDAEBITDAEBITDA (Earnings Before Interest, Taxes, Depreciation, and Amortization) measures a company's operating profitability before financing and accounting decisions, used to compare core performance across firms.View full definition → to cure a breach), asset sales, or a repayment to reduce net debt.
3. The waiver or amendment — approaching lenders *before* the breach to reset the covenant or obtain a waiver.
The cardinal rule: never surprise your lenders. A covenant breach discovered by a bank in your quarterly compliance certificate is a crisis. The same breach flagged by the CFO eight weeks early, with a credible remediation plan, is a manageable amendment. The difference is entirely in the timing and the relationship—which brings us to the third pillar.
Knowledge check
1. What is the central lesson CFOs should draw from how the SVB collapse affected its corporate borrowers?
2. Why does the lesson describe a single term loan as a 'blunt instrument' compared with a tranched capital structure?
3. When would a CFO prefer a Term Loan B over a Term Loan A for a particular financing need?
4. Select ALL correct answers about the characteristics of a revolving credit facility (RCF) as described in the lesson.
Select all the correct answers.
5. Select ALL correct answers about how the lesson frames the CFO's approach to debt facilities and banking relationships.
Select all the correct answers.
Credit is a relationship business dressed up as a transaction business. The CFO who treats banks as vendors—running every mandate as a competitive auction to shave fees—will find the phone unanswered in a downturn. The CFO who treats the banking group as a managed portfolio of relationships will find capital available precisely when the market is closed to everyone else.
Understand the economics from the bank's side. The RCF is often a loss-leader—banks provide undrawn commitment capacity at thin returns because it wins them the profitable ancillary business: cash management, FX, hedging, trade finance, and the fees from your next bond issue or M&A financing. This is the share-of-wallet bargain, and it is entirely explicit in a modern relationship.
The strategic error is failing to allocate that wallet deliberately. If a bank commits $100 million to your revolver and receives none of your FX or cash-management flow, that relationship is quietly deprioritized inside the bank—and when you need an amendment, your file lands with a credit officer who has no reason to fight for you. The CFO's job is to maintain a wallet-allocation model: which banks hold which commitments, and what ancillary revenue each receives in return. Keep it roughly balanced, and keep your key relationship banks *profitable* on the total relationship.
Concentration is risk—as SVB's borrowers learned. A single-bank relationship is efficient until the day that bank retrenches, gets acquired, exits your sector, or fails. The judgement is to build a syndicate large enough to diversify counterparty risk but small enough to coordinate in a crisis. For a mid-cap company, a core group of three to five relationship banks is typically the sweet spot: enough redundancy that any one bank's withdrawal is absorbable, few enough that you can get them all on a call in a week.
Structure this deliberately. Designate one or two banks as lead relationships—they get the largest commitments, the most ancillary business, and the closest access. The others are participants, sized to fill out the facility and provide diversification. In distress, your leads become your advocates inside their own credit committees.
The relationship is built in calm markets and drawn upon in stressed ones. The operating discipline:
The test of a banking relationship is not the price it gives you at signing. It is whether, when your covenant runway shortens to two quarters and the credit markets have seized, your lead banker walks into their credit committee and argues *for* you. That advocacy is earned over years of transparency and wallet, and it is the single most valuable and least visible asset on your funding balance sheet.
1. Engineer facilities across tenor, tranching, and optionality—not just price. Match each instrument to the asset it funds and the investor base that prices it best; build accordion capacity and negotiate a *narrow* MAC clause while your credit is clean, because you cannot negotiate flexibility in a crisis.
2. Negotiate the EBITDA definition, not just the covenant level. Add-backs, synergy credits, and pro-forma treatment can move your leverage ratio by half a turn—a generously defined 4.0x can beat a narrowly defined 4.5x. The definitions are where control is won or lost.
3. Know your covenant runway cold and manage to a cushion. Run 15–20% inside every maintenance covenant, model the exact quarter and scenario in which you breach, and act through the lever hierarchy—operational, then structural, then amendment—*before* the tripwire is hit.
4. Never let a lender learn of a problem from a compliance certificate. The same breach is a crisis if discovered and a routine amendment if telegraphed eight weeks early with a remediation plan. Timing and relationship convert distress into a manageable conversation.
5. Manage the banking group as a wallet-allocated portfolio. Keep three to five relationship banks, allocate ancillary business to keep each core relationship profitable, diversify against counterparty failure, and invest in the cadence—so that when markets close, your lead banker argues your case inside their own credit committee.