Interview preparation
What real estate private equity interviews test, with short answers, one case solved to the dollar and the structure of a timed modeling test.
A real estate private equity interview usually tests three things: whether you can explain the vocabulary without notes, whether you can do the arithmetic of a deal on paper, and whether you can build a clean model under time pressure. This page gives you ten questions with the answers interviewers expect, one debt sizing case worked in full, and the structure of a typical modeling test. Free, no sign-up.
Answer each one in under a minute: the direct answer first, then one number, then one nuance.
Start with sources and uses: price, closing costs and loan fees on one side, the loan and the equity cheque on the other. Then the operating cash flow (NOI, less capital items, less debt service), the exit (forward NOI over an exit cap, less selling costs, less the loan), and finally levered and unlevered IRR and multiple. Finish with the two or three assumptions the answer depends on most, usually rent growth and the exit cap.
The risk-free rate, plus a risk premium for the asset (location, tenant, lease length, liquidity), minus expected NOI growth. As a rule of thumb, a cap rate is roughly the required return minus growth, which is why low-growth assets trade at higher cap rates.
Value falls by 0.5 / 5.5, or 9.1%, with NOI unchanged. The loan does not move, so the equity, 40% of value, absorbs the whole loss: 9.1% / 40% = 22.7%. That is the arithmetic of leverage, and interviewers want to hear the second number.
Below NOI: capital expenditure, tenant improvements, leasing commissions and reserves, then debt service. Two buildings with the same NOI can produce very different cash flows, which is why a cap rate alone never prices a deal.
Yes, when the cost of debt is above the unlevered return. A building bought at 100 that yields 6 and sells at 100 returns 6.0% unlevered. Finance 60% of it at 8% interest-only and the equity return falls to 3.0%. That is negative leverage: more risk for less return.
Debt yield is NOI divided by the loan. It ignores the interest rate, the amortisation and the appraiser’s cap rate, so it tells the lender what the loan earns if it had to take the keys tomorrow. It is often the binding test when cap rates are low.
About 15%: 1.15 to the fifth is roughly 2.01. The exact figure is 14.87%. Two more worth knowing cold: 3.0x in seven years is about 17% (16.99%), and money doubles at 12% in about six years (rule of 72; exactly 6.12).
Cash goes first to return the partners’ capital, then to an 8% preferred return, often pro rata. Above the pref, the sponsor takes a promote: a larger share of each further dollar than its share of the equity, sometimes in tiers (for example above an 8% and then a 12% IRR hurdle). Some structures add a catch-up that brings the sponsor up to its full share of profit before the split resumes.
Usually not. Most underwriting adds 25 to 50 basis points: the building is older at exit, its lease profile is shorter and the plan should not depend on market timing. Then show the exit cap at which the deal still clears its target return.
Compare the yield on cost with the market cap rate. A plan that stabilises at 6.5 of NOI on 100 of total cost yields 6.5% on cost; if comparable stabilised assets trade at 5.25%, the finished building is worth about 123.8, a 23.8% margin on cost. Then test the rents against recent leases, the downtime, the capex per square foot and the exit cap that wipes the margin out.
Debt sizing comes up in almost every REPE case interview, on paper or in Excel. This one is Case 8 of the REPE Interview Prep Pack, shortened, with every figure recomputed by an independent script.
You are arranging a five-year fixed-rate loan on a stabilised grocery-anchored retail centre. The lender sizes the loan as the lowest of three tests.
| Input | Value |
|---|---|
| Net operating income | $5.0M |
| Appraised cap rate | 6.25% |
| Interest rate, fixed | 6.50% |
| Amortisation, monthly payments | 30 years |
| Maximum loan to value (LTV) | 65% |
| Minimum debt service coverage (DSCR) | 1.25x |
| Minimum debt yield | 10.0% |
Questions: what is the maximum loan, which test binds, and what changes if the loan is interest-only or the rate is 7.50%?
Value = $5.0M / 6.25% = $80.0M.
The DSCR test needs the annual debt service per dollar of loan, the mortgage constant. With monthly payments, r = 6.50% / 12 and n = 360: payment per dollar = r / (1 − (1 + r)−n), times 12 = 7.585% a year. On paper, at rates of 6% to 7% a 30-year constant sits roughly one point above the rate (here 1.08 points).
| Test | Formula | Loan |
|---|---|---|
| LTV | 65% × $80.0M | $52.00M |
| DSCR | $5.0M / (1.25 × 7.585%) | $52.74M |
| Debt yield | $5.0M / 10.0% | $50.00M |
| Maximum loan | lowest of the three | $50.00M |
The debt yield test binds at $50.00M. At that loan the LTV is 62.5%, the DSCR 1.32x and the debt yield exactly 10.0%.
Interest-only. With a 6.50% constant, the DSCR test would allow $5.0M / (1.25 × 6.50%) = $61.54M. The loan does not move: the debt yield test ignores the rate and the amortisation, so it stays at $50.00M. That is precisely why lenders added it.
Rate at 7.50%. The constant rises to 8.391% and the DSCR loan falls to $47.67M. The maximum loan becomes $47.67M: the DSCR test now binds.
The likely follow-up question: which test binds where? When cap rates are low, values are high relative to income, so debt yield tends to bind. When interest rates are high, DSCR binds. LTV binds when cap rates are high relative to the lender’s yield requirement.
To run the same arithmetic on your own numbers, the free loan sizing template puts the four tests side by side.
Modeling tests come in three common formats: a paper test (30 to 60 minutes, no computer, the arithmetic above), a timed Excel test (usually one to three hours, a short case memo and a blank workbook), and a take-home case (a day or more, a model plus a short investment memo or a few slides). The timed version is the one candidates find hardest, and it almost always asks for the same building blocks, in this order.
An illustrative time budget for a three-hour test: about 15 minutes reading and setting up inputs, 90 minutes on the pro forma, debt and returns, 30 minutes on the waterfall and sensitivities, and the last 30 minutes kept for checks and the written answer. Candidates more often fail on an unfinished model than on a wrong formula, so a complete simple model beats an elegant half-built one.
A paid companion for candidates who want more cases like the one above. What is in the download:
Inputs in blue, every calculation a live formula, built-in checks, no macros, no locked cells. Each model is pre-filled with its matching case. Written for analyst and associate candidates who already know what NOI, a cap rate and an IRR are. All cases are fictitious and simplified; this is educational material, not investment advice. Single-user licence.
€99, instant download, sold on Gumroad.
See the REPE Interview Prep PackThe three books below cover the same ground at full length. Each has a free companion page with its workbooks.