The technical set: where the weight actually is

01where the weight is

What is asked most, what depends on your background, and how long it takes to prepare.

• Accounting first, and out loud^1,2

  The three statements and how they link are the base of everything else, and the classic test is a single change - a factory bought with debt, say - walked through all three. Both main guides put accounting at the front of the queue. Practise it as speech, because that is how it will be asked.

• Valuation, and the trade-offs between methods^1,2

  Most technical questions are on basic accounting and valuation, with the depth scaled to your background: a finance master's student gets harder valuation questions than a history graduate. Know why you would use one multiple over another - EV/EBITDA, for instance, when you want to strip out capital expenditure entirely.

• Accounting and valuation are everyone's, whichever group you want³

  Accounting and valuation run through every corporate finance team, so a candidate for ECM, DCM or restructuring is expected to handle them as well as an M&A candidate. The reverse also holds at a basic level: an M&A candidate should know enough about debt to discuss how an acquisition is financed, because the leverage in a buyout is what drives its return. The division pages set out what each product group adds on top.

• Questions from outside your group stay basic⁴

  One way to budget the time: from a division you are not applying to, expect only the elementary questions - what a bond is, what an IPO does - and treat anything deeper as optional. The difficulty that is aimed at you rises with your background; a second internship earns harder questions than a first.

• Start two to three months out¹

  For someone without a finance background, the recommended runway for the technical set is two to three months, against a day or two for fit. Brain teasers are explicitly not worth the time.

next: accounting →

02accounting

• Link the statements, not the debits and credits^1,2

  Nobody will ask for journal entries; everyone will ask how the three statements connect. Net income at the foot of the income statement is the top line of the cash flow statement; non-cash items such as depreciation are added back and working-capital movements adjusted for; investing and financing flows follow; and the ending cash becomes the balance sheet's cash, while net income flows into retained earnings. If assets do not then equal liabilities plus equity, a link is wrong.

• A worked change: an asset sold below its book value¹

  A factory carried at 100 is sold for 80, at a 40% tax rate. The income statement books a loss of 20, so net income falls by 12. The cash flow statement starts from that 12 lower, adds the 20 loss back because it is non-cash, and shows the 80 of proceeds under investing: cash is up 88. On the balance sheet cash rises 88 and property falls 100, so assets are down 12 - matched by retained earnings, down 12. Every single-change question is this pattern.

• Working capital, said in one sentence¹

  The change in working capital says whether a company has to spend ahead of its growth or gets paid ahead of it. A retailer buys inventory before it sells, so the change is usually negative; a subscription business collects cash in advance, deferred revenue rises, and the change is often positive. It feeds free cash flow, and so the valuation.

next: valuation and the dcf →

03valuation and the dcf

• Enterprise value and equity value¹

  Equity value is what the business is worth to shareholders alone; enterprise value is what it is worth to every provider of capital. The questions test whether you can move between the two and say what a given transaction does to each, not whether you can recite the bridge.

• Choose a multiple for a reason¹

  EV/EBITDA strips out capital expenditure, depreciation and capital structure entirely; EV/EBIT strips out capital structure but keeps depreciation in, which suits industries where equipment is what drives value, manufacturing for instance. P/E is distorted by tax rates, leverage and non-core income, and is still the one to use for banks and insurers, where interest is part of the business rather than its financing.

• Which method comes out highest¹

  There is no fixed answer, and saying so is the first half of a good one. The second half: precedent transactions usually come out above trading comparables, because a buyer pays a control premium that a share price does not include, while a DCF gives the most variable result because it rests on assumptions far into the future.

• The DCF, as an extension of one formula^1,2

  A DCF is the present-value formula extended over an explicit forecast period plus a terminal value. Expect to explain each input and which of them move the answer most; the bank here has questions on exactly that.

• Walk through a DCF in five steps^1,5

  Forecast free cash flow to the firm for five to ten years. Estimate a terminal value for everything after, by a perpetuity growth rate or an exit multiple. Discount both to today at the weighted average cost of capital, which gives enterprise value. Subtract net debt and other non-equity claims to reach equity value, and divide by the share count. Then show a sensitivity table, because the answer moves a great deal with its assumptions.

• Why WACC falls and then rises as debt is added^1,5

  Debt is cheaper than equity, partly because interest is tax-deductible, so moving from no debt to some lowers the weighted cost of capital. Past some point the risk of default lifts the cost of both debt and equity and the average turns back up. Asked what happens when leverage goes from 20% to 30%, the honest answer is that it depends where on that curve the company sits - probably still falling at those levels.

• Check the terminal value against itself^1,5

  Whichever method you use, back into the other. An exit multiple implies a perpetual growth rate, and a growth rate implies a multiple; an implied growth of 10% a year for a company in a developed economy means the multiple is too high. The usual perpetuity growth range is 1% to 3%, around expected inflation.

next: m&a and lbo →

04m&a and lbo

• M&A and LBO: conceptual at entry level^1,6

  Merger and buyout models are described as less important than accounting, valuation and the DCF for an entry-level banking interview; you need to reason about accretion, dilution and leverage, not model them. The paper LBO - IRR and multiple in five to ten minutes with pen and paper - is primarily a private equity screening tool, so it matters most when you interview for the exit, not the entry.

• Accretion or dilution from two P/E multiples¹

  In an all-stock deal, compare the buyer's cost of paying in shares - one over its own P/E - with the yield it buys, one over the price it pays for the target's earnings. A buyer on 25 times earnings paying 15 times pays 4% for a 6.7% yield, so the deal adds to earnings per share. With cash or debt in the mix the shortcut stops working and you need the interest figures.

• Paper-LBO arithmetic: two anchors^1,6

  Doubling your money over five years is roughly a 15% IRR; tripling it is roughly 25%. With those two points you can place most returns in your head: buy 100 of EBITDA at 10 times with 60% debt, sell 150 of EBITDA at 9 times after repaying 250 of the debt, and the equity goes from 400 to 1,000 - 2.5 times, about 20%.

• What makes a good buyout candidate¹

  Price first: almost any company works at a low enough price, and a high one raises the chance of failure sharply. Then stable, predictable cash flow, no heavy ongoing capital spending, and a realistic exit - with the return coming from EBITDA growth and debt repayment rather than from hoping to sell at a higher multiple.

↑ back to top