A development facility is not a big mortgage
The single most expensive misunderstanding in development finance is treating the loan like a home loan that happens to be larger. It is not. A home loan is advanced once and repaid over years. A development facility is a limit you draw against progressively, as the project consumes it, and repay in one hit from settlements.
Three consequences follow, and all three show up in the numbers below. Interest accrues on the balance outstanding rather than the facility, so a slower start costs less than a slow finish. Interest is usually capitalised — added to the balance rather than paid monthly — so the debt grows even while nothing else happens. And because it is capitalised, the facility has to be large enough to hold it.
The four tests, and why “we lend to 65%” means nothing on its own
Debt as a share of total development cost. A test lenders commonly apply alongside the next one, and a useful one to know — but it is not the basis of the figures on this page. Sixty to sixty-five per cent is a common band for residential development; it is not a rule.
The loan against the gross realisation on completion, net of GST, rather than against cost. This is the one the figures on this page are built on: the calculator caps the loan at the LVR times gross realisation ex GST, and the interest and fees capitalised into the loan count inside that cap. A deal can pass one test and fail the other, which is why being told "we lend to 65%" tells you almost nothing until you ask 65% of what.
Net realisation divided by total debt. It answers the only question the lender truly cares about: if this sells at less than you think, does the loan still come back?
Qualifying presales as a multiple of debt. Often the condition that decides whether the facility is drawn at all, and the one most likely to move the programme.
A percentage without its denominator is not a term, it is a mood. Sixty-five per cent of cost and sixty-five per cent of end value are different amounts of money on the same project, and on a thin deal they are different answers.
Peak debt is bigger than your debt, and the gap is the loan itself
On the worked deal below, the model reports a debt figure of $4,228,068 — the part of the development cost the loan funds. But the loan’s balance at its peak is $4,529,318, which is 65.0% of the $6,968,182 gross realisation ex GST — the facility limit.
It is the interest and the establishment fees, capitalised into the balance: $4,228,068 + $237,829 + $63,421 = $4,529,318. The loan borrows to pay for itself. A feasibility that quotes only the debt figure understates the balance the facility has to carry by $301,250 on a deal this size — and because that balance has to stay inside the lender’s LVR limit, the interest and fees also take room away from the cost the loan can fund.
This is why the equity figure matters more than the debt figure. Equity here is the cost the loan does not fund — $5,656,650 − $4,228,068 = $1,428,582 — where the first figure is total development cost before finance, because the finance is capitalised into the loan rather than paid in cash. It is the number that has to be real, in cash, before anything is drawn.
The leverage trade, in actual numbers
The same six-townhouse deal used in the feasibility study guide — $1,650,000 land, 6 dwellings at $1,250,000, $520,000 a dwelling to build, 18 months in Victoria — run at 6 different gearing levels. Everything else held constant, including the calculator’s land-tax holding estimate ($10,500 a year, an estimate that also grows with the programme; council rates and utilities are not estimated). The loan is sized on LVR against gross realisation ex GST, with the interest and fees capitalised inside the limit.
| LVR | Debt | Peak loan | Equity required | Interest | Net profit | Margin on cost |
|---|---|---|---|---|---|---|
| 50.0% | $3,252,360 | $3,484,091 | $2,404,290 | $182,945 | $1,079,801 | 18.3% |
| 55.0% | $3,577,596 | $3,832,500 | $2,079,054 | $201,240 | $1,056,628 | 17.9% |
| 60.0% | $3,902,832 | $4,180,909 | $1,753,818 | $219,534 | $1,033,455 | 17.4% |
| 65.0% | $4,228,068 | $4,529,318 | $1,428,582 | $237,829 | $1,010,282 | 17.0% |
| 70.0% | $4,553,304 | $4,877,727 | $1,103,346 | $256,123 | $987,109 | 16.5% |
| 75.0% | $4,878,540 | $5,226,136 | $778,110 | $274,418 | $963,936 | 16.1% |
Gearing up makes the project look worse. From 50.0% to 75.0%, net profit falls by $115,865 and margin on cost drops 2.3pt, because every extra dollar borrowed adds interest, and interest is a cost like any other.
But look at the equity column. It falls from $2,404,290 to $778,110 — a fall of 67.6%. You gave up 10.7% of the profit to cut the cash you tie up by that much.
That is the entire argument for leverage, and it is invisible to anyone reading margin alone. It is also why a developer and their financier can look at one feasibility and reasonably disagree about whether the gearing is right.
We have deliberately not printed a return-on-equity percentage in that table, for the same reason the study guide does not: the denominator is a genuine choice — peak equity, average equity over the programme, or total cash outlay — and each gives a different answer on identical inputs. The two engine figures are there side by side. Anyone quoting you an ROE without saying which denominator they used has skipped the only part that made it meaningful.
What the rate actually costs
Same deal, gearing held at 65.0%, interest rate moved 4 points across a range that has been ordinary in Australia within the last few years. The peak loan does not move — it is the LVR limit — so a higher rate raises the interest and leaves the cost the loan can fund slightly smaller.
| Rate | Interest | Net profit | Margin on cost |
|---|---|---|---|
| 5.50% | $176,885 | $1,070,326 | 18.1% |
| 6.50% | $207,572 | $1,040,092 | 17.6% |
| 7.50% | $237,829 | $1,010,282 | 17.0% |
| 8.50% | $267,665 | $980,886 | 16.4% |
| 9.50% | $297,090 | $951,897 | 15.8% |
4 percentage points of rate cost $118,429 of profit and 2.3pt of margin. Real, and worth negotiating — but hold that figure next to the one in the next section before deciding where to spend your effort.
Time is the real interest rate
Same deal, same 65.0% and 7.50%. Only the programme changes — and with it the interest and the land-tax holding estimate, both of which run for the whole programme.
| Months | Interest | Holding (land tax) | Net profit | Margin on cost |
|---|---|---|---|---|
| 12 | $161,377 | $10,500 | $1,090,854 | 18.6% |
| 15 | $199,940 | $13,125 | $1,050,235 | 17.8% |
| 18 | $237,829 | $15,750 | $1,010,282 | 17.0% |
| 24 | $311,650 | $21,000 | $932,301 | 15.4% |
| 30 | $382,975 | $26,250 | $856,781 | 14.0% |
Going from 12 months to 30 costs $234,073 of profit and 4.5pt of margin — about 2.0 times the $118,429 that a 4-point rate rise costs. Put per unit: a point of rate is about $29,607 of profit on this deal, a month of programme about $13,004.
Developers negotiate rates hard and programmes softly. The numbers say a long delay is the bigger exposure. A month of planning delay is not an inconvenience with a cost attached later; it is a cost, immediately, and it compounds.
It is also the most common modelling error in the whole exercise. Interest computed on a twelve-month term for an eighteen-month build understates cost and overstates profit quietly, because nothing in the output says which term was used.
What to bring to a lender
| A feasibility whose assumptions are stated and dated, not a single profit figure. |
| A programme the interest was actually computed over, matching the facility term you are asking for. |
| The planning position: permit in hand, lodged, or not yet, and what that does to the timeline. |
| Cost evidence proportionate to the ask — a builder's letter is not a fixed-price contract, and lenders know the difference. |
| Price evidence for the revenue line, because it is the assumption with the most power over the answer and usually the least support. |
| Your equity, and where it is. Cash already spent on land is equity; cash you intend to raise is not. |
Not thin margins — unexplained ones. A lender is underwriting your assumptions, not your optimism, and the fastest way to fail is a feasibility that produces a confident number nobody in the room can trace back to a source. That is a tooling question as much as a diligence one.
Put your own gearing through it
The calculator runs the same engine that produced every table on this page. Change the LVR, the rate and the programme and watch which one actually moves your answer.