Covenant Reset: What Lenders Ask For in Return

A financial covenant is set at the start of a loan against a forecast. When the business moves away from that forecast and is not going to move back within a quarter or two, every future test date becomes a breach waiting to be certified. Excusing them one at a time, with a waiver at each quarter end, wastes everyone's time and tells the lender nothing new. A reset rewrites the levels for the test dates ahead, so the covenant measures the business as it now is and can warn again if it slips further.

Resets are asked for after a breach and, more usefully, before one. A request that arrives with a forecast before the test date shows a finance team reading its own numbers; a breach the lender discovers from a compliance certificate shows the opposite. The amendment can also come at the borrower's year end for an accounting reason. Under IFRS and the national standards based on it, a loan in breach at the balance date can have to be shown as current even if the lender waives the breach before the accounts are signed. An agreement reached before the balance date can avoid that, one reason such requests often arrive just before it; debt classification on a breach sets out the effect.

The amendment names new levels for specified test dates, often stepping back towards the original level as the forecast recovers, and says when the original covenant returns. Some resets suspend a test for a period instead; some change the definition behind the ratio, such as an EBITDA add-back, which can loosen a covenant as effectively as a new number. A lender reads all three as the same request.

The price is negotiated and varies by deal. Lenders commonly ask for an amendment fee and a higher margin, sometimes only until the ratio returns to the original level. Alongside the money come protections: monthly management accounts in place of quarterly, a thirteen-week cash flow, a block on dividends and shareholder loan repayments, tighter permitted baskets, an equity injection, security over assets that were free, or a shorter maturity that brings the conversation forward. In a bilateral facility the bank alone decides. In a syndicated or club deal, covenant levels usually go to the majority lenders as the agreement defines them, and fees are often paid only to lenders who consent.

A commercial bakery, constructed for this page like every figure in it, has a leverage covenant of 3.00 times tested quarterly. At December it reports debt of US$41m against twelve-month EBITDA of US$13.0m, which is 3.15 times. Flour and energy costs have risen faster than its contracts let it pass on, and its forecast shows recovery over a year as price reviews come through. It asks for a reset. The levels below are the bank's proposal in this example, not a market norm.

The last column is the test of the proposal. If the bakery's March EBITDA comes in more than 12.5% below forecast, it breaches; a smaller miss does not. Now set the levels as a borrower under pressure might ask, each just above the forecast: 3.30, 3.15, 2.95 and 2.70 times. The same column reads 0.6%, 0.8%, 2.8% and 2.9%. A covenant with that little room is breached by an ordinary miss, and the next request arrives in three months. The bank agrees the first schedule, with a fee, a margin step-up until leverage is back under 3.00 times, monthly reporting and no distributions until then.

DEFINED IN ONE PARAGRAPH → · THE EXAM HALL QUESTION ON IT → · LOAN COVENANTS, EXPLAINED →

A waiver looks backwards: it is the lender's agreement not to act on a particular breach. A reset looks forwards, and the two often travel together, the waiver for the test already failed and the reset for those to come. An equity cure is the borrower's own route out of a single breach, paid for in new money and limited in how often it can be used. A covenant holiday suspends testing altogether for a period, which removes the warning until it ends. Each answers a different question, and a lender asked for one is entitled to ask why the others would not do.

The reset decides whether the covenant goes on doing its job. A lender reading a file looks at the history of the levels as well as today's ratio. One reset, set against a credible forecast and followed by recovery, is a covenant working as intended: it caught the problem early and the parties agreed a path. A run of resets, each placed just above the latest result, is a covenant that has stopped warning anyone, and a lender that finds that pattern has learned that the forecasts behind each reset were wrong. The question then is no longer where to set the line. It is whether the business can carry its debt, and the file has moved towards problem loans and workout.