The EBITDA bridge. Why an unqualified audit opinion does not protect the sale price of a company

I took part in a meeting that summed up the financial due diligence of a company. At the meeting, the buyer’s advisers presented four EBITDA adjustments. The seller’s model did not include any of them. Together they reduced the company’s result by about PLN 2 million. In three cases, the buyer’s advisers were right. The fourth adjustment also changed working capital, that is, inventory and receivables less trade payables. In doing so, it undermined the working capital price adjustment mechanism that the parties had negotiated two months earlier.

The seller’s financial statements carried an unqualified audit opinion. The starting figures in the model agreed with the audited financial statements. On Friday afternoon, the walk from EBITDA in the financial statements to adjusted EBITDA looked reasonable. On Wednesday afternoon, the buyer’s advisers spent three hours going through the model line by line and arrived at a different result.

Between the signing of the letter of intent and closing, the seller lost about 18 percent of the price stated in the letter. Most of that amount disappeared that Wednesday afternoon. The price is usually calculated as a multiple of EBITDA, so every zloty taken off the result lowers it by several zlotys.

I have changed the details so that neither the company nor the parties can be identified.

What an EBITDA bridge is

EBITDA is operating profit plus depreciation and amortization. The buyer treats it as a measure of how much the company earns from its core business. It most often sets enterprise value, that is, the value of the business itself, excluding cash and debt, as a multiple of EBITDA.

The result in the financial statements, however, is rarely the figure from which the price is calculated. The seller adjusts it for events that will not recur and for costs that will disappear or change after the transaction. The table that shows the path from one figure to the other is called an EBITDA bridge in deal jargon. The first row shows EBITDA from the audited income statement. Each subsequent row is one adjustment, with an amount and a justification. The last row shows adjusted EBITDA, from which the buyer calculates enterprise value.

Alongside the bridge, the model has a second table. It shows the step from enterprise value to the price for the shares, that is, the amount that will reach the seller’s account. In this table, net debt, meaning bank loans, other borrowings and lease liabilities less cash, is deducted from enterprise value. The result is then adjusted for the difference between working capital on the closing date and the level agreed in the contract.

Adjustments in the two tables affect the price to different degrees. Every zloty removed from adjusted EBITDA lowers enterprise value by as many zlotys as the multiple. Every zloty added to net debt or missing from working capital lowers the price for the shares by one zloty. An adjustment that reduces EBITDA and at the same time increases net debt or the working capital shortfall on the closing date therefore lowers the price in two places: once through the multiple, and a second time zloty for zloty.

What the auditor does not check

An unqualified audit opinion gives the seller the sense that their figures will hold up in any review. The auditor, however, answers a different question. The auditor assesses whether the financial statements as a whole give a true and fair view of the company’s position in accordance with the Polish Accounting Act. Minor discrepancies that do not change the picture given by the financial statements as a whole do not affect the opinion. The adjustments that the seller adds to the result for the purposes of the transaction are not assessed by the auditor at all, because they are not in the financial statements.

The quality of earnings review, QoE for short, is commissioned by the buyer from its financial advisers after the letter of intent is signed. The review ends before the sale agreement is signed. Closing, that is, payment and the transfer of the shares, takes place on the day the agreement is signed or later, once its conditions are met, for example when the competition authority gives its approval. The price in the letter of intent is therefore a conditional price. The buyer states it on the assumption that the seller’s figures will be confirmed and checks them only during exclusivity.

In this review, the result is checked for each month separately. In this way, the buyer assesses the trend, and it calculates working capital at the end of each month. It also compares management reports with the audited financial statements. In many mid-sized companies, the books are kept mainly for tax purposes and for the annual financial statements. Bonuses, customer rebates, provisions and inventory write-downs are booked once a year, in December. For eleven months, the monthly data therefore show a higher result and lower liabilities than the true position. The auditor checks whether revenue and costs have been recorded in the right period, but does so at the year end. At month ends during the year, only the buyer’s advisers check this.

Where the seller’s model stops holding up

The places where the seller’s model stops holding up are predictable and almost always the same.

The first of them is normalizations, that is, adjustments meant to show how much the company would earn on market terms, without the owner and without one-off events. I described the full list of such adjustments in my article on the quality of earnings review. Owners often get the direction of one of them wrong. If the owner pays himself twice the market salary of a CEO, the excess is added back to EBITDA. If he works for a token amount, the buyer will deduct from EBITDA the cost of a CEO employed on market terms. Each normalization is checked separately, and each has to be backed by a document.

The second place is the quality of revenue. This is where sellers in mid-sized transactions usually lose the most. The income statement shows total sales. The customer structure, the length of contracts and the risk of losing customers have to be read from other data. The buyer’s advisers therefore break revenue down by individual customer and by month. They check what share of sales comes from the five largest customers and when their contracts expire. They also count how many customers have left. Growth in orders from two large customers can hide the departure of a dozen or so smaller ones.

Customer concentration rarely changes the bridge itself. In that case, the buyer lowers the multiple or moves part of the payment into a contingent payment, that is, an earn-out, or into an escrow account, where the money stays blocked for an agreed period. The loss of a major customer before the transaction, however, goes straight into the bridge. The buyer then calculates how much the company would earn without that customer and removes from EBITDA the margin the customer brought in. Sellers do not usually add such a row themselves. All of these findings appear in the buyer’s advisers’ report two weeks after the seller considered the price agreed.

The third place is working capital. The sale agreement sets its reference level, that is, the amount that under normal conditions has to remain in the company on the closing date. If there is less working capital at that point, the price falls by the difference, zloty for zloty. I explained why a twelve-month average fails in a seasonal company in my article on the locked box and closing accounts.

The buyer also changes the composition of working capital itself. It excludes from the average the months in which the main customer paid much earlier or much later than usual. It moves overdue trade payables into net debt and takes them out of working capital. It often treats payables under reverse factoring in the same way. In that arrangement, the bank pays the supplier on the company’s behalf, and the company repays the bank later. The seller treats such liabilities as ordinary trade payables, and the buyer as debt that will have to be repaid. A favorably negotiated reference level then often loses its meaning, because the buyer recalculates it from scratch.

The least obvious is the link between an EBITDA adjustment and working capital. Many entries in the income statement also change working capital items on the balance sheet. If August revenue has to be moved to September because the service was performed only in the following month, receivables at the end of August also fall. If the costs of services performed in August but invoiced in September have to be added to August, payables at the end of August rise. Each such change shifts historical working capital, and with it the reference level calculated on the old figures. The fourth adjustment described at the start worked in exactly this way.

Why three items cost the whole price

If the bridge leads, for example, to PLN 10 million of EBITDA and three of its rows do not agree with the books, the buyer assumes that the others are also questionable. The renegotiation then covers the whole price. From that moment, the buyer’s advisers look for further errors in every item. The lack of control over the figures itself becomes a risk, which the buyer prices in through a lower multiple or by moving part of the price into a deferred payment.

The price in the letter of intent was approved by the buyer’s investment committee or management board on the basis of EBITDA from the seller’s bridge. A team that goes back to the committee with three incorrect items has to defend the whole figure from scratch. It is easier for that team to propose a lower price. I described the role of such a team and its sponsor in a separate article.

At this point, the seller is in its weakest negotiating position. The other bidders were told long ago that talks had been put on hold, and exclusivity does not allow the seller to go back to them.

Two versions of the bridge

Almost no seller’s model in a mid-sized transaction passes this review without corrections. The reason is simple. The seller’s adviser or the company’s CFO builds the model for the information memorandum, the document in which the company presents itself to buyers. The memorandum is meant to show the company in the best light, so every doubtful item is resolved in the seller’s favor. The owner grows attached to the valuation that results. No one on the seller’s side checks the model the way the buyer’s advisers will.

The solution is simple and uncomfortable. From day one, two versions of the bridge have to be built. The first goes into the memorandum. The second shows the owner which adjustments the buyer’s advisers will challenge, which documents are missing and how much of the price will be left after due diligence. It is best to do this a year before the process, because then an incorrect item can still be corrected in the books. During exclusivity, it is the buyer that prices it.

The second version starts from the trial balance, a list of all accounts in the company’s books with their balances. Each row of the bridge leads to a specific account, then to the entries in the general ledger and to the source document, that is, an invoice, a contract or a payroll. Data from three years is also attached to each row, because the buyer will check whether the event recurred in the past. An adjustment based on a management estimate along the lines of “some of these costs were one-off” is usually rejected. Monthly management reports have to add up, to the zloty, to the result in the annual financial statements. Revenue has to be broken down by customer and month, and working capital calculated at the end of each month.

It is worth writing the definitions of EBITDA, net debt and working capital into the letter of intent, as I wrote in the context of exclusivity. During due diligence, the buyer then checks the figures according to rules both sides know. At Capital One Advisers, we build both versions of the bridge before the first page of the memorandum is written.

A well-built bridge will not stop a buyer that wants to lower the price for other reasons. The buyer’s advisers will examine the company themselves anyway, even if they receive a report commissioned by the seller. In a smaller transaction, the cost of such a report may exceed the amounts in dispute. Some adjustments will remain contentious, because they depend on judgment. No one knows whether a saving introduced in October will hold for the whole year, or whether a customer that cut back its orders will again buy as much as before. Preparation mainly changes the moment at which the seller learns about the problem.

The bottom line

The seller described at the start lost about 18 percent of the price because of figures that had been wrong for six months. No one had checked them earlier the way a buyer checks them. At Capital One Advisers, the owner sees both figures, his own and the one the buyer will calculate, before anyone on the other side signs a confidentiality agreement.

One question remains open, and sell-side advisers themselves disagree about it. Some of them leave a few bold adjustments in the bridge so that they have something to give up in negotiations. They expect such adjustments to raise the price in the letter of intent while buyers are still competing with one another. Others show only adjustments for which they can defend every zloty. They believe that once exclusivity has been signed, every adjustment that does not survive due diligence will cost the seller more credibility than it earlier added to the price. Who is right depends on how strong the competition is on the day the letter of intent is signed and how many of these adjustments the seller can document.

Service: M&A advisory on a company sale