Inspecting Free Cash Flow to the Firm

The third bridge ends with a business whose cash a buyer can model, and a list of what it produced to prove it. The fourth bridge begins by checking that list, line for line, each line with a bar set by the people who will rely on the figure next: a lender's credit analyst, and the modelling team of an equity buyer standing behind that lender.

The check comes first because the fourth bridge sizes debt on free cash flow to the firm. A lender who lends against a figure that later proves wrong loses money, and a lender who suspects the figure lends less. Inspecting it as rigorously as a lender will, before any debt is raised against it, means the bridge starts from cash a credit committee would accept.

Who the inspectors are

Professionals inspect much of the list. A quality-of-earnings review covers the cash and most of the forecast, an appraiser or surveyor the assets, the outside accountant the inventory count, a tax adviser the tax and counsel the structure. An independent model reviewer takes the model, the rate and the scenarios, with a lender's or buyer's own model as the outside check. Where the lead inspects, the lead works from their findings and does not stand in for them.

Is the cash real?

The first six lines check that the cash exists and is measured on a consistent basis.

Capital spending has to cover thirty-six months under a written capitalization policy the outside accountant has approved, with every item above the policy's floor matched to an invoice and classified as maintenance or growth. Average annual maintenance spending is then set beside depreciation. The two should sit reasonably close, and where they do not, the gap is explained asset class by asset class. Nobody expects them to match, since depreciation describes what was bought years ago and maintenance describes what it costs to keep it running now. A large gap can be fine; a gap nobody can explain is a finding. In the illustration, maintenance of $410,000 a year against depreciation of $500,000 is 82 percent of it, and the $90,000 between them is what the asset classes have to account for.

The fixed-asset register has to tie to the ledger with no more than a trivial difference, every entry sighted within the last year, and a qualified third party has to have rated the condition and remaining life of the assets that make up most of the replacement value. There must be a five-year replacement plan at current cost, and deferred maintenance either done or priced as a deduction. A lender cares because equipment that fails during a loan is paid for out of the cash meant to service it.

Working capital has to cover twenty-four month-ends on one written definition that excludes cash, debt and debt-like items, with a peg at the trailing twelve-month average and every adjustment documented. In nearly every month of the latest year, net working capital should have landed close to what the season would predict, and receivable, payable and inventory days should have held to target through recent months. The inspector is testing control as much as level.

The balance sheet has to be collectible, saleable and complete. Receivables long past due should be a small share of the total, reserved under a written policy drawn from three years of write-offs. A full inventory count within the last year, observed by the outside accountant, should agree closely with the books, with slow-moving stock reserved. And a written revenue recognition policy must match every deferred balance and deposit to the work still owed.

Every obligation a buyer will deduct from enterprise value has to be on the schedule of debt-like items and quantified, from accrued bonuses and customer deposits to earn-outs, leases that behave like debt and warranty exposure. Here the bar is absolute: nothing found in diligence may be missing from the schedule. On this bridge it matters twice, because a lender counts the same items as claims ranking ahead of the owners.

The cash itself has to be visible. That means a monthly cash flow statement reconciled to the bank for twenty-four months, and a thirteen-week forecast rolled weekly for long enough to have a record, in which the forecast made four weeks earlier lands close to actual closing cash in the great majority of weeks. Cash conversion must be reported for three years, with every material movement explained.

What is the business, and what tax does it bear?

The next two lines check what is being valued.

The tax memo has to reconcile three years of cash taxes at the level of the entity to the returns and to the book provision, with every difference scheduled. NOPAT is computed at a stated rate, effective or notional, with its basis written down. The study of where the business owes tax covers every state with sales, staff, contractors or property: the business is registered and filing wherever it has to be, past exposure is settled or quantified and reserved, and no audit is open without a reserve.

The structure has to put every controlled entity that deals with the business on one chart, with its ownership, tax classification and holdings. Every payment between entities sits under a written agreement at a documented market rate, with balances reconciling at every close. And a written perimeter of sale must hold everything the operating business needs, or an assignable lease or licence over it. A lender will take security over what is inside the perimeter, so the perimeter is a question of collateral as well as of value.

Can the future be forecast?

Five lines check the forecast, because a lender sizes debt on the future.

The budgets have to cover the last two years and the current one, by month, built by the leads and approved before each year began, and reaching down to capital spending and working capital. Each of the last two should have landed close to actual at EBITDA. Two budgets are two observations, which warrant little on their own, so the inspector checks that the forecast was pulled toward the base rate for businesses of this size in proportion to how short the record is. Every material line variance must have been explained in writing for twenty-four months.

The model has to be an integrated five-year forecast of all three statements, whose first year is the approved budget and which reproduces the last two years closely at EBITDA when run on their actual drivers. Revenue is built from the customer schedule and retention history, and growth does not run ahead of the business's own recent record without a documented reason. The DCF at the cost of capital from the valuation file is the governing figure. The multiple is computed beside it as pricing, the gap between them is explained, and the gap is never closed by adjusting the DCF. Sensitivities and an independent review, with every finding closed, complete the line.

The reinvestment register has to cover three years of growth capital spending and growth working capital by project, each with the revenue and operating profit it produced. The return on reinvested capital is computed on a stated definition and sits above the cost of capital, with the number of projects stated and the return carried into the forecast shrunk toward the industry base rate accordingly. Forecast growth can be no higher than the reinvestment rate multiplied by that return, and after the convergence year the return on new capital equals the cost of capital.

The incremental margin, the change in operating profit divided by the change in revenue, is shown by line for each of the last three years and reconciled to the fixed and variable cost split the budget uses. Any growth year whose incremental margin fell below the reported margin is explained.

And the growth period has to be stated: the year growth settles to a mature rate no higher than the risk-free rate, an advantage period no longer than the base rate for businesses of this size, and a return on new capital that fades in a straight line to the cost of capital. A forecast that breaks any of these is a finding, whether it grows above the mature rate for too long, carries an excess return past the convergence year, or sets the mature rate or the advantage period beyond those limits.

What is it worth?

Eight lines check the valuation file, because the equity value the fourth bridge computes is built on it.

The return on total invested capital has to cover three years on the one definition the framework has used since the first anchor, decomposed into margin and capital turnover and set against the weighted cost of capital, with the register's marginal return beside it and the difference explained. The competitor map has to name every competitor in each served market with three years of share from a stated source, evidence a barrier through customer captivity, a cost advantage or local scale, and derive an advantage period that agrees with the convergence year in the forecast. A map with no named competitor is a finding, and so is a convergence year the map does not support.

The reproduction-value schedule values every asset, tangible and intangible, at what it would cost to reproduce, each with its source. The cost of capital file shows every component with its source and date. It carries no premium for size or for the business's particular risks; each of those risks is named at the cash-flow line or scenario where it is priced, and the rate is not lowered by assuming more debt than the business can carry. On this bridge that last condition is about to be tested for real.

The scenario file places every forecast against base rates for businesses of this size, carries three scenarios with probabilities drawn from the business's own history, and states the margin of safety by layer. The price reversal log shows every offer, broker's opinion and comparable transaction the business has met, reversed into what it implies, with no input changed afterwards without a recorded reason. The holder's inputs are stated on the seller's own capital structure, never a buyer's. And growth is split into organic, pricing and acquisition, each with its own return, with any acquisition earning below the cost of capital carried as negative growth value.

Did the crossing pay?

The last line is the owner's. For the crossing just completed, the holder's return is the cash the owner took out plus the change in equity value from the previous anchor's conclusion to this one, divided by the equity value at the start. It is computed at the same level of value at both ends and set against the relevered cost of equity over the same period. Beside it sits the history of distributions: what was paid in each year, and what could have been. A return below what the owner required is reported as such and not explained away, and the owner decides, with that number in hand, whether the next bridge is worth crossing.

Where a failure goes, and how a business slips

A business that fails a line goes back to the third bridge, to the first span whose measure it cannot show. A capital spending record that does not hold goes back to the first span, a working-capital peg that cannot be delivered to the second, a tax position with open exposure to the third, and a forecast that cannot reproduce its own history to the fourth.

The inspection also lists how a business slips off free cash flow to the firm after reaching it, and the list reads as the third bridge's disciplines lapsing one at a time. On the cash side, maintenance is deferred, the capitalization policy bends to flatter a year, the asset register goes stale, working capital balloons, debt-like items accumulate off the schedule and the cash forecast stops being updated. Tax exposure builds in new states, and assets and contracts drift outside the perimeter of sale. Forecasts miss without explanation. Growth spending stops being tracked to what it earned, growth is bought with margin, and forecast growth runs past the year it was meant to settle. In the valuation file, the competitor map goes stale without the advantage period being re-derived, the cost of capital is not refreshed and drifts from the market, the scenarios collapse back to one case, and an acquisition is credited with the organic return and never measured on its own capital. Last, the holder's return goes unmeasured, and the owner never learns whether the crossing paid.

Once there is debt, each lapse costs more. On the third bridge it cost the business some of its value; on the fourth, the same lapse can cost it a covenant.