Closing the Gap – Part 5: The Holdback

Earlier in this series

• Part 1 – Why the Gap Is About Risk, Not Value
• Part 2 – Why Earn-Outs Break Down and What Protects Your Payment
• Part 3 – Vendor Loans, and What It Costs to Finance Your Own Sale
• Part 4 – The Retained Stake, and What a Second Bite Really Costs

Money parked against something specific that might go wrong. Of the five structures, it is the most defensible, and the easiest to let sprawl into something else.

The Holdback In Brief

A holdback keeps part of the price back against identified risks, commonly five to fifteen per cent of the price for twelve to twenty-four months.

How is it different? The other four bridge a disagreement about value. This one covers a named contingency: a disputed tax position, a customer contract that needs consent to assign, a warranty that might be called on.

Where does it go wrong? Scope and release. A holdback against two named risks becomes a general escrow with an open tail, and a release that requires the buyer to agree is a discount wearing a different word.

What is the test? It should shrink as diligence progresses. If it is still large at signing, diligence has not done its job.

The holdback is the narrowest of the five and the one most sellers accept without argument. That is reasonable. A buyer who has found a real contingency is entitled to cover it, and a seller who refuses looks like someone with something to hide.

The argument is not about whether. It is about what, how much, how long, where the money sits, and what has to happen for it to come back.

Name the risks, or don’t hold the money

A holdback is only as good as the list it is held against. Each risk should be named, with its own cap and its own release date. A single pool covering “any claim under the agreement” is not a holdback, it is a reduction in price that you may or may not recover.

Two numbers do most of the work. A de minimis stops trivial claims being made at all, and a basket means small claims have to accumulate past a threshold before any of them can be brought. Without both, you will spend a year arguing about amounts that cost more to argue about than to pay.

Where the money sits for a holdback

If the holdback stays in the buyer’s account, you are an unsecured creditor of the person who owes you money and decides whether to pay it. Put it with a third party under an escrow deed, or hold it in a bank account with joint signatories.

Then make release automatic. The default should be that the money comes to you on the release date unless the buyer has made a claim in the proper form before it. Anything else means you spend the last month of the period chasing a counterparty with no reason to hurry.

Disputes go to an independent expert, named by role and appointed by a professional body if the parties cannot agree. Courts are slower and more expensive than the amounts usually in dispute.

Match the clock to the risk

One release date for everything is lazy drafting and it always favours the buyer. A customer consent either arrives or it does not, and ninety days will tell you. A warranty claim on the accounts follows the review period. A contested tax position runs as long as the Commissioner’s amendment period, which can be considerably longer than the deal team assumed.

Set each clock to the risk it covers. The effect is that most of the money comes back early, and only the genuinely long-tailed piece stays out.

Watch the stacking

If the deal also carries a vendor loan or an earn-out, check how the set-off provisions interact. Buyers routinely take the right to deduct claims from the loan balance and to draw on the holdback, which lets the same dollar of alleged loss be recovered twice. Say so in the agreement: claims come out of the holdback first, and only the excess reaches anything else.

A holdback that is properly named, capped, timed and escrowed is the cheapest bridge of the five, because you are not carrying the buyer’s uncertainty, only a risk that both of you have already identified and sized.