The letter of intent is signed. The purchase price looks clean. Then, somewhere between diligence and closing, a number appears that nobody adequately explained during negotiations — the working capital peg — and it quietly carves six, seven, sometimes eight figures off the wire transfer. This is not a rare edge case. It is the most consistent source of seller disappointment in middle-market M&A, and it is almost entirely preventable if addressed before ink touches the LOI.

What the Peg Actually Is

Every acquisition involving a going concern carries an implicit assumption: the buyer expects to receive a business with enough current assets to run operations from day one. The working capital peg is the agreed target level of net working capital — typically current assets minus current liabilities — that must be present at close. If actual working capital falls short, the purchase price adjusts downward dollar-for-dollar. If it exceeds the peg, the seller captures the surplus.

Buyers almost always propose the peg. Their preferred methodology is a trailing twelve-month average of net working capital, often calculated from management accounts the seller provided during diligence. The structural problem: that average includes seasonal peaks, unusually strong collection periods, or one-time inventory builds — levels the seller cannot reliably replicate in the closing month. The peg gets set high; the closing balance comes in lower; the adjustment flows to the buyer.

Where the Negotiation Gets Expensive

Three specific levers determine whether the peg favors the seller or the buyer. First, the calculation methodology — whether the peg is a simple average, a median, or a normalized figure excluding anomalies. Sellers with cyclical revenue or lumpy receivables should push hard for a normalized approach that strips outlier months. Second, the line-item inclusions — which accounts actually flow into the computation. Deferred revenue, customer deposits, and accrued liabilities are frequently contested; their treatment can shift the peg by several hundred basis points of enterprise value. Third, the measurement date — closing adjustments are calculated on estimated figures first, then trued up post-close, sometimes 90 to 120 days later. The post-close true-up is where disputes concentrate, and where escrow dollars sit exposed the longest.

Sellers who do not retain a quality-of-earnings advisor before signing the LOI are effectively letting the buyer’s team define all three levers uncontested. That is the sequence that produces surprises.

Pre-LOI Positioning That Changes the Outcome

The LOI is the moment of maximum seller leverage, and most sellers spend it negotiating headline price while leaving peg methodology entirely to the definitive agreement. That is the wrong sequence. Before the LOI is signed, sellers can:

• Run their own trailing working capital analysis across 18 to 24 months, identifying the defensible normalized range and the months they would exclude as non-representative.
• Propose a peg range with a deadband — a corridor of, say, plus or minus a defined dollar threshold — within which no adjustment triggers. This eliminates small disputes that nonetheless consume legal and management time.
• Define in the LOI term sheet which line items are explicitly included or excluded, rather than leaving it to “customary definitions” that the buyer’s counsel will draft.
• Negotiate the escrow percentage and true-up timeline simultaneously with the peg itself, since these three terms are structurally interdependent.

The Operator Read

The working capital peg is not a technicality — it is a pricing mechanism dressed in accounting language. Buyers understand this. Sophisticated sell-side operators treat the peg as a negotiating variable with the same discipline they bring to EBITDA multiples. The structural dynamic is straightforward: whichever party defines the methodology first tends to hold the advantage at closing. Sellers who wait for the purchase agreement draft to engage on this point are negotiating after the leverage has already shifted.