45 technical questions, with the answers

The income statement shows profitability over a period: revenue down through costs to net income. The balance sheet is a position at a single date: assets equal liabilities plus equity. The cash flow statement reconciles the two, taking net income and adjusting for non-cash items and changes in working capital to explain the movement in cash between two balance sheets. They link: net income flows into retained earnings and starts the cash flow statement, and the closing cash balance is the cash line on the balance sheet.

The cash flow statement. Value is the present value of cash a business generates, and the cash flow statement is the only one of the three that reports cash directly rather than after accrual judgements. It also contains most of the other two: it begins with net income from the income statement and moves through the working capital and capex lines that drive the balance sheet.

Income statement: operating income falls $10, tax falls $2.50, net income falls $7.50. Cash flow statement: start with net income down $7.50, add back the $10 of non-cash depreciation, so cash is up $2.50. Balance sheet: cash up $2.50 and PP&E down $10, so assets are down $7.50; retained earnings are down $7.50 by the fall in net income. It balances.

Earnings before interest, tax, depreciation and amortisation. It is used because it strips out financing decisions, tax jurisdiction and the accounting treatment of past capital spending, which makes two companies with different capital structures and asset bases roughly comparable. It is also a rough and ready proxy for operating cash generation before working capital and capex.

Operating working capital is receivables plus inventory less payables: the cash tied up in running the business day to day. An increase is a use of cash: the company has funded more receivables or inventory, or paid its suppliers faster. That is why it appears as a negative in the cash flow statement even though the balance sheet asset went up.

On day one: cash up £1,200, deferred revenue up £1,200 as a liability. Nothing touches the income statement, because nothing has been earned. Each month, £100 moves out of the liability and into revenue, with the associated tax. By month twelve the liability is zero and £1,200 of revenue has been recognised, with no further cash movement.

Income statement: a $100 write-down reduces pre-tax income by $100, tax falls $25, net income falls $75. Cash flow: net income down $75, add back the $100 non-cash write-down, so cash is up $25. Balance sheet: inventory down $100, cash up $25, so assets down $75; retained earnings down $75. It balances.

Goodwill is the residual in purchase accounting: the price paid for a business less the fair value of its identifiable net assets, after those assets have been written up or down and any separately identifiable intangibles have been recognised. It is created only in an acquisition. It is not amortised under either IFRS or US GAAP; it is tested for impairment, and written down when the acquired business is worth less than its carrying value.

A write-up of assets to fair value that is recognised for book purposes but not for tax. The acquirer will report higher depreciation or amortisation in the accounts than it can deduct on its tax return, so book tax expense is lower than cash tax paid over the life of the asset. The difference is recognised at close as a deferred tax liability, calculated as the write-up multiplied by the tax rate, and it unwinds as the asset is depreciated.

Not necessarily. Negative book equity arises from accumulated losses, but just as often from large buybacks funded with debt, or from a spin-off or dividend recapitalisation. A highly profitable business that has bought back more stock than its historical retained earnings will show negative equity while generating substantial cash. The question is whether the company can service its obligations, which is answered by cash flow and coverage, not by a book figure that reflects historical cost.

Equity value is what belongs to shareholders: diluted shares times price. Enterprise value is the value of the operating business itself, before deciding who funds it. The bridge is enterprise value = equity value + debt + preferred + minority interest − cash. Cash is subtracted because an acquirer effectively receives it back on day one, so the true cost of owning the operating business is lower by that amount.

Comparable companies, precedent transactions and a discounted cash flow. Precedents usually give the highest value because transaction prices include a control premium, where trading comps reflect minority stakes. The DCF is the wildcard: it can land anywhere, because it is driven entirely by the assumptions you feed it. In a frothy market comps can trade above historical precedents, so "usually" is doing real work in that sentence.

Because EV/EBITDA is neutral to capital structure and to tax jurisdiction, so it compares two businesses financed differently on the same basis. P/E is an equity metric: it reflects leverage, so a highly levered company will show a lower P/E for reasons that have nothing to do with the quality of its operations. EV/EBITDA is also unaffected by differences in depreciation policy, which matters when comparing companies with different asset ages.

Last full fiscal year, plus the nine months just reported, less the nine months from the same period a year earlier. So FY2025 plus 9M 2026 less 9M 2025, which gives the twelve months to September 2026. Every figure is reported, so nothing in it is an estimate. Before quoting a multiple on it, check whether the company bought or sold anything inside that window, because the raw arithmetic would then be blending two different companies and needs a pro forma adjustment.

Calendarise it. Weight each fiscal year by the number of months it contributes to the calendar year: for a 31 May year end put onto calendar 2026, five twelfths of the year ending May 2026 and seven twelfths of the year ending May 2027. Apply the weights to income-statement and cash-flow lines only, never to the balance sheet, which is a position at a date, so you take the reported date closest to the one you are aligning to. Then say on the page that the figures are calendarised, because a calendarised and an uncalendarised page look identical.

They fix different problems. LTM moves one company’s window forward to the most recent twelve months it has reported, which solves staleness: four months after a year end, the last annual figure is already old. Calendarisation aligns several companies with different year ends onto a common one, which solves comparability. LTM is exact, because it is arithmetic on reported periods: last full year, plus the interim just reported, less the same interim a year earlier. Calendarisation is an approximation, because it weights two fiscal years by months of overlap and assumes trading is spread evenly across the year. A comps page often needs both: LTM for the trailing column, calendarisation for the forward one.

On business model first, not on sector label: what the company sells, to whom, with what growth, what margin structure and what capital intensity. Then screen for size within an order of magnitude, and for geography where regulation or end markets differ materially. A set of six close comparables is far more useful than twenty loose ones, and you should be able to say in one sentence why each name is in it, and, when asked, why an obvious name is not.

Find out why before deciding anything. Check first for a mechanical explanation: a depressed EBITDA from a one-off, a pending bid in the price, a different fiscal year not yet calendarised, a mis-set net debt. If the multiple is real, it is usually growth, margin or market position, in which case the company may not belong in the set, or belongs in it with the reason stated. Present the median and the mean, show the range, and footnote the outlier rather than silently deleting it.

Because it produces the maximum a financial buyer could pay while still hitting its return threshold, and a financial buyer captures no synergies. A strategic acquirer with cost or revenue synergies can rationally pay more, so the LBO value tends to sit at the bottom of the football field. It is a floor only in a market where sponsors are active bidders; in a distressed market with no available financing, it is not a floor at all.

Move up the income statement until you reach a line that is positive and meaningful: EV/Gross Profit or EV/Revenue, cross-checked against where the business should be on a Rule of 40 style measure. Then value it on what it will look like: a DCF built to a normalised margin, with the assumption stated explicitly and sensitised, or a multiple applied to a forward year in which the business is profitable, discounted back. Sector-specific measures matter too, such as EV per subscriber, per bed or per megawatt, because they price the asset rather than the accounting.

Project unlevered free cash flow for five to ten years: EBIT, less tax on EBIT, plus depreciation and amortisation, less capex, less the increase in net working capital. Discount each year at WACC. Calculate a terminal value, either by perpetuity growth or by an exit multiple, and discount that back too. Sum to enterprise value. Bridge to equity value by subtracting net debt, preferred and minority interest, then divide by diluted shares for a value per share. Present it as a sensitivity table across WACC and terminal growth, never as a single number.

The blended return the company must earn to satisfy everyone funding it: the cost of equity weighted by the market value of equity, plus the after-tax cost of debt weighted by the market value of debt. Cost of equity comes from CAPM: the risk-free rate plus levered beta times the equity risk premium, with any country or size premium the context demands. Cost of debt is what the company would pay to borrow today, read off its traded bonds or built from its credit rating, and it enters after tax because interest is deductible. The weights are market values, never book.

Two methods. Perpetuity growth: final-year free cash flow times one plus g, divided by WACC minus g. With $500m of final-year FCF, 2.5% growth and a 9% WACC, that is $500m × 1.025 ÷ 0.065 = $7.9bn. Or an exit multiple applied to final-year EBITDA, cross-checked against where comparable companies trade. It matters because it is typically 60 to 75% of total enterprise value on a ten-year forecast. A perpetuity captures an infinite series where the explicit period captures a decade. Best practice is to run both and check they agree; if they diverge widely, one of the assumptions is wrong.

The risk-free rate rises, so the cost of equity rises and the cost of debt rises, so WACC rises and the value falls. The effect is largest for companies whose value sits furthest in the future, meaning high-growth businesses with most of their cash flow in the terminal value, which is the mechanism behind long-duration equities selling off hardest when rates move. If the rate rise reflects higher expected inflation rather than higher real rates, nominal cash flows should also rise, which partly offsets it. Saying that second part is what separates the answer.

When cash flows cannot be forecast with any confidence, or when the discount rate is not meaningful. Early-stage companies with no revenue history, commodity producers whose cash flow is a function of an unforecastable price, and banks and insurers are the standard three. For banks, debt is raw material rather than financing, so unlevered free cash flow and WACC do not mean what they mean elsewhere. For banks you would use a dividend discount model or a return-on-equity to price-to-book framework instead.

Standard discounting assumes cash arrives in a lump at each year end. In reality it arrives through the year, so the mid-year convention discounts each year’s cash flow at the half-year point, at period 0.5, 1.5 and 2.5, which raises the value by a few percent. Use it where cash flow genuinely is spread through the year, which is most operating businesses. The thing that catches people out is applying it inconsistently: if the forecast period uses mid-year, the terminal value has to be discounted on the same basis.

Accretive. The acquirer is issuing paper valued at 20x to buy earnings priced at 10x, so each new share brings in more earnings than it dilutes. The shortcut for all-stock deals is to compare the two P/E ratios after the premium: higher acquirer P/E means accretive. That is before synergies, and before any change in the financing mix, both of which move the answer.

Compare the target’s earnings yield, the inverse of its P/E, against the after-tax cost of the cash used. If the company draws on debt at 6% and pays 25% tax, the after-tax cost is 4.5%. A target bought at 10x has a 10% earnings yield, so the deal is accretive by a wide margin. If the cash comes off the balance sheet, the cost is the after-tax interest income forgone, which is usually much lower and makes almost any cash deal accretive on paper. That last point is why accretion is a weak test of whether a deal is good.

Cost synergies come from removing duplicated functions, consolidating sites and buying better, and they are the ones with a credible track record: they are within the acquirer’s control and they can be planned. Revenue synergies come from cross-selling and pricing, and they depend on customers behaving as hoped, which they frequently do not. Financial synergies, such as a lower cost of capital or a better tax position, are real but rarely the reason for a deal. The professional position is to underwrite cost synergies with a defined plan and a cost to achieve, and to treat revenue synergies as upside rather than as part of the price.

When its own shares are richly valued, when the deal is too large to fund with cash or debt, when it wants the target’s shareholders to share the integration risk, or when it needs to preserve balance sheet capacity. The signalling problem is well documented: issuing stock tells the market management thinks the shares are expensive, which is why acquirers’ shares often fall on announcement of an all-stock deal. Target shareholders may also prefer stock for tax reasons, since a stock-for-stock exchange can defer a capital gain that a cash sale crystallises.

No. Accretion is an arithmetic consequence of the relative multiples and the funding mix, and it says nothing about whether value was created. A company can buy a declining business with cheap earnings and print EPS accretion in year one while destroying value, because it has bought a lower growth rate and a worse risk profile. The tests that matter are whether the return on invested capital exceeds the cost of capital, whether the synergies are real and achievable at the cost assumed, and whether the acquirer can integrate it.

Start with the purchase price. Write the target’s assets and liabilities up or down to fair value, and recognise separately identifiable intangibles that were not on its balance sheet: customer relationships, technology, brands. Where an asset is written up for book but not for tax, recognise a deferred tax liability equal to the write-up times the tax rate. Whatever remains between the price and the fair value of identifiable net assets is goodwill. The identifiable intangibles are then amortised over their useful lives, which is why an acquirer’s reported earnings fall even when the business performs as underwritten.

A sponsor buys a company using a mix of debt and equity. Build sources and uses: the purchase price plus fees is the use, funded by debt tranches and a sponsor equity cheque. Project the business, run the debt schedule with mandatory amortisation and a cash sweep so free cash flow repays debt each year, and roll the balance sheet. At exit, apply an exit multiple to the final year’s EBITDA to get enterprise value, subtract the remaining net debt to get equity value to the sponsor, and compute the multiple of money and the IRR. Returns come from three places: EBITDA growth, debt paydown and multiple expansion.

Predictable, cash-generative earnings that can service debt through a cycle. Low capital intensity, so cash flow converts rather than being reinvested to stand still. A defensible market position that limits the chance of a sudden revenue decline. Assets that can support secured lending. A clear path to improvement through cost, pricing or bolt-ons, so the return does not rely on multiple expansion. And an identifiable exit: a strategic buyer who would want it, or a public market that would take it.

MOIC is total cash out divided by cash in, and it ignores time entirely. IRR is the annualised rate that discounts the cash flows to zero, so it is highly sensitive to timing. A 2.0x in three years is a far better outcome than a 2.0x in seven, and only IRR shows it. That sensitivity is also why IRR can be gamed: an early dividend recapitalisation or a subscription line that delays the capital call raises IRR without changing the cash a limited partner ultimately receives. This is why LPs look at both, together with DPI.

Entry enterprise value $1,000m, debt $500m, so equity in is $500m. At exit, enterprise value is 10 × $150m = $1,500m, debt is $500m − $200m = $300m, so equity out is $1,200m. MOIC is 1,200 ÷ 500 = 2.4x. Over five years that is roughly a 19% IRR, since 2.4x over five years sits between the 2.0x that is about 15% and the 2.5x that is about 20%.

What lenders will underwrite, which comes down to cash flow and asset coverage rather than to what the sponsor wants. The binding constraints are the interest coverage the credit committee requires, the maximum total leverage the market will clear at the time, and the stability of the cash flow through a downside case. A stable software business with high recurring revenue supports more turns than a cyclical manufacturer at the same EBITDA. Market conditions move this year to year by several turns, which is why the same deal is financeable in one year and not the next.

Decompose the change in equity value into three pieces. EBITDA growth: entry multiple applied to the change in EBITDA. Multiple expansion: exit EBITDA times the change in multiple. Debt paydown: the reduction in net debt over the hold. They sum to the change in equity value, and running the bridge tells you what you are really underwriting. A return that depends mostly on multiple expansion is a bet on the market, not on the business, and a committee will say so.

The most senior piece of the capital structure that does not get repaid in full at the enterprise’s current value, so it is the layer where value breaks. In a restructuring, everything above it is paid out and everything below is wiped out, and the fulcrum typically converts to the equity of the reorganised company. Identifying it is the central analytical act in distressed investing: you value the enterprise, walk down the structure until the money runs out, and buy the security that will own the business.

The bid is the highest price a buyer will pay, the offer the lowest a seller will accept; you buy at the offer and sell at the bid, so the spread is the cost of a round trip. It widens when the risk of quoting rises, which is two risks: inventory risk, that the position moves before it can be hedged, and adverse selection, that the counterparty knows something. Thin volume, higher volatility, a pending announcement and smaller companies all raise one or both, which is why spreads blow out in a crisis.

Estimate a central case out loud: roughly 65 million people, most driving, a station serving a few thousand cars, giving an order of magnitude in the thousands rather than the hundreds of thousands. Then quote a spread that reflects how uncertain that estimate is: something like 6,000 at 10,000, wide because the estimate is weak. If the interviewer lifts your offer, you are short and you should move your market up before quoting again. What is graded is whether the spread reflects your uncertainty and whether you update after being traded against.

Short-dated yields above long-dated ones means the market expects rates to be cut, which usually means it expects the economy to weaken. In the United States inversion has preceded every recession since the 1960s, but at lead times running from about six months to two years, so it is a statement about direction rather than timing. The more actionable observation is that the curve typically re-steepens before the recession begins, so the shape of the disinversion carries more information than the inversion itself: bull steepening from the front end rather than bear steepening from the long end.

Each contract is on 100 shares, so the position is 200 × 100 × 0.35 = 7,000 deltas short. Buy 7,000 shares and you are neutral to a small move. When the stock rises to $52, delta rises, say to 0.45, so the position is now 9,000 short and the hedge is 2,000 shares light. Being short options is being short gamma: rebalancing forces you to buy into strength and sell into weakness, and that is the cost of the premium you collected.

No, it has deteriorated. The spread widened from 500 to 550 basis points. The yield only fell because the risk-free rate fell further. This is why credit is quoted, traded and risk-managed on spread rather than on yield: the spread isolates compensation for credit risk from the level of rates.

Three things that matter. Rate: loans are floating, bonds are usually fixed, so bonds carry interest-rate risk and loans do not. Amortisation: loans amortise, bonds are bullet. Covenants: loans carry maintenance covenants tested every quarter, bonds carry incurrence covenants tested only when the company takes an action. Loans also generally sit senior and secured, which is why they recover more in a default and why they are more often the fulcrum in a restructuring.