Macquarie IB Interview Questions

Yes. A toll road in India would typically have a higher expected IRR (internal rate of return, the yearly return an investment's cash flows produce) than one in Indiana. Investors want to be paid for the extra risks: currency, political and regulatory, construction, and traffic-forecast risk. In the riskier market they pay less for the same expected cash flows, and that pushes the expected IRR up.

For a quick bank valuation, I'd want tangible book value (equity minus goodwill and intangibles), forward return on tangible common equity (ROTCE), asset quality, capital strength, and the price-to-tangible-book (P/TBV) multiples of comparable banks. Expected performance matters most, because a price pays for future profits. The past is useful mainly as evidence of whether the forecast is believable.

Together they paint the house in 1.2 hours, which is 1 hour 12 minutes. Jimmy paints 1/3 of a house per hour. John paints 1/2 of a house per hour. Combined, that is 1/3 + 1/2 = 5/6 of a house per hour, so one house takes 6/5 hours.

Assuming a standard $1,000 face value and annual compounding, the bond is worth about $857.34 today: $1,000 ÷ (1.08)². A zero coupon bond pays no interest along the way. It pays only its face value (the amount repaid at maturity), so the price is that one payment discounted back two years at 8% compounded.

A project finance model values one asset, such as a toll road or a wind farm, owned by a special purpose vehicle (SPV: a company set up only to hold that project). Its lenders are repaid mainly from the project's own cash flows, not from a parent company's balance sheet. The model runs over a fixed life in two phases. Construction covers the capital cost and how debt and equity fund it. In operations, revenue, often contracted, less operating costs, taxes, working capital changes and maintenance capex gives cash flow available for debt service (CFADS). From CFADS you size the debt to a target debt service coverage ratio (DSCR: cash flow divided by the interest and principal due), run the cash through a waterfall that pays lenders before owners, and read off the equity IRR (the owners' annual return) and the coverage ratios lenders test.

Revenue is traffic multiplied by toll rate, and I forecast the two separately over the life of the concession, the fixed-term right a government grants a private operator to run the road and collect tolls. I build volume by vehicle class and grow it off a macro driver such as regional GDP, population or employment. A newly opened road carries a multi-year ramp-up, and an elasticity assumption takes out the traffic lost when tolls rise faster than inflation. Toll rates escalate under the formula written into the concession agreement, most often CPI or a contractual floor. The product then runs through a collection ratio to allow for discounts, exemptions and unpaid trips.

An investor would typically target a levered equity IRR of roughly 8%-12% on a mature toll road that is already open and has a traffic history, and roughly 13%-18% on a greenfield road they have to build first. IRR is the annualised return an investment earns over its life. Levered means it is measured on the equity cheque after debt service, so debt raised against the concession lifts the return as long as the asset's unlevered return beats the cost of that debt. The gap between the two ranges is the price of construction and ramp-up risk. On a greenfield asset you are underwriting a traffic forecast that has never been tested, and you want several extra points of return for it.

Project the road's cash flows year by year across the remaining concession life, the fixed term of usually 25 to 50 years during which the operator has the right to run the road and keep the tolls, then discount them at a low WACC. Revenue is traffic volume times the average toll, with tolls escalating at CPI or whatever formula the contract allows; from that you deduct operating cost, lifecycle capex (the periodic resurfacing and structural work a road needs every decade or so), and tax. Leave out the Gordon Growth perpetuity and the exit multiple. At expiry the asset hands back to the grantor, so the model simply runs out of years, and the only terminal item is a contracted handback or residual payment if the concession provides one.