Ask most people how a rental makes money and they will tell you about the rent. Ask them how to evaluate one and they will tell you about cash flow.
Both answers are incomplete in the same way. A leveraged rental property returns money through three separate channels, and cash flow is the only one that shows up on a monthly statement. The other two — the loan being retired by someone else's rent, and whatever the property is worth when you sell — are frequently larger, and they are invisible to cap rate, cash-on-cash, and DSCR alike.
Here is what that costs in practice. Two identical properties, differing only in down payment. One earns $345 a month and one loses $12 a month. Over ten years with no appreciation whatsoever, the first returns 29.1% and the second loses money. Add 3% annual appreciation and the ranking reverses: the property with negative cash flow returns 131.9% against the other's 102%.
Neither result is a trick. They are the same arithmetic run against different assumptions about the one thing nobody can predict.
A note before you start. This is general education, not investment advice, and nothing here is a recommendation about any property, market, or capital structure. The worked comparison uses illustrative inputs stated on the page: a $265,000 property with $17,000 of net operating income, financed at 7.125% over 30 years, held ten years, with 8% selling costs. Appreciation figures cited as ranges are FHFA-derived state-level measurements from this site's own sourced dataset — they describe what already happened, and applying them forward is a scenario, not a forecast. Tax effects are deliberately excluded throughout, for the reason given in section 7.
1. The three channels
Cash flow is rent minus operating expenses minus debt service. It arrives monthly and it is the only channel you can spend.
Principal paydown is the portion of each mortgage payment that retires the loan. It is real equity, accruing every month, funded by the tenant. You cannot spend it until you sell or refinance, which is why it feels less real than it is.
Appreciation is the change in the property's value. Historically the largest component on a leveraged property, and entirely dependent on something outside your control.
There is a fourth thing people sometimes list — tax benefits — and this article deliberately excludes it. That is covered separately in how rental property is taxed, because mixing pre-tax and post-tax figures in one analysis is how people end up double-counting depreciation.
2. Principal paydown, and why it accelerates
This is the channel most consistently underestimated, and the reason is that its shape is counterintuitive.
Every fixed-rate mortgage payment is the same amount, but the split inside it moves. Early on, most of it is interest, because the balance is large. Late on, most of it is principal. That means paydown starts small and grows every single year, for the same payment.
On a $212,000 loan at 7.125% over 30 years:
| Year | Principal retired that year |
|---|---|
| 1 | $2,102 |
| 5 | $2,793 |
| 10 | $3,984 |
| 15 | $5,683 |
| 20 | $8,107 |
Year twenty retires nearly four times what year one did, from an identical payment. That is compounding running in your favour, and it is why the case for a rental strengthens with time in a way that a cash-flow analysis of year one cannot show.
The corollary matters as much: a short hold captures very little of it. Five years into a 30-year loan you have retired about 5.7% of the balance. Anyone planning a three-to-five-year hold should be honest that paydown will contribute little, and that the case therefore rests on cash flow and appreciation alone.
3. The worked comparison
Same property, same NOI, same rate. Only the down payment differs.
Property A — 40% down. $106,000 down, $159,000 loan, annual debt service $12,855. Property B — 20% down. $53,000 down, $212,000 loan, annual debt service $17,139.
Both with $9,000 of closing costs and initial work, so cash invested is $115,000 and $62,000 respectively.
| A (40% down) | B (20% down) | |
|---|---|---|
| Annual cash flow | +$4,145 | −$139 |
| Monthly | +$345 | −$12 |
| Cash-on-cash | +3.60% | −0.22% |
On cash flow alone, A wins and B is a deal most people would reject.
Now run both for ten years and look at the whole return.
With no appreciation at all:
| A | B | |
|---|---|---|
| Cash flow collected | $41,450 | −$1,390 |
| Principal paydown | $22,160 | $29,546 |
| Appreciation | $0 | $0 |
| Total profit | +$33,414 | −$2,048 |
| Return on cash invested | +29.1% | −3.3% |
| Annualised | 2.6% | −0.3% |
A still wins, comfortably. B loses money over a decade.
With 3% annual appreciation:
| A | B | |
|---|---|---|
| Total profit | +$117,261 | +$81,799 |
| Return on cash invested | 102.0% | 131.9% |
| Annualised | 7.3% | 8.8% |
The ranking flips on the percentage return. B produces less profit in absolute dollars and a substantially higher return on the capital committed, because it tied up $53,000 less of it.
Run your own deal at three appreciation rates4. What that flip actually means
It is tempting to read the second table as an argument for leverage. It is not. It is an argument for understanding what leverage does.
Leverage amplifies the appreciation, in both directions. Property B controls the same $265,000 asset with $53,000 rather than $106,000. Every dollar the property gains or loses in value accrues to a smaller capital base, so the percentage swing is larger either way. At 3% appreciation that works out well. At 0% it does not. At negative appreciation it is considerably worse than these tables show.
Which structure wins depends entirely on the least knowable input. That is the honest summary. A analyses better on every metric that uses only known quantities — cash flow, cash-on-cash, DSCR — and B wins only in scenarios where the property appreciates. Choosing B is choosing to be right about appreciation.
Absolute profit and percentage return are different questions. A earns $35,462 more in actual dollars at 3% appreciation. B earns a higher percentage on a smaller commitment. Which matters depends on whether your constraint is capital or opportunities — an investor with $115,000 and one deal should think differently from one with $115,000 and two deals available.
5. Where the break-even sits
The comparison in section 3 shows two endpoints. The more useful question is where the line between them actually falls — at what rate of appreciation does the lower-cash-flow structure overtake the higher one?
Running the same ten-year hold at a range of rates:
| Annual appreciation | A profit (40% down) | B profit (20% down) | A return | B return |
|---|---|---|---|---|
| −3% | −$30,602 | −$66,064 | −26.6% | −106.6% |
| 0% | $33,414 | −$2,048 | 29.1% | −3.3% |
| 1% | $58,921 | $23,459 | 51.2% | 37.8% |
| 2% | $86,805 | $51,343 | 75.5% | 82.8% |
| 3% | $117,261 | $81,799 | 102.0% | 131.9% |
| 5% | $186,739 | $151,277 | 162.4% | 244.0% |
Read the two right-hand columns rather than the profit columns, because those are what compare structures fairly.
The crossover sits somewhere between 1% and 2% annual appreciation. Below it, the more conservative structure wins on return as well as on cash flow. Above it, leverage takes over and the gap widens quickly.
Two things about that band are worth sitting with.
It is uncomfortably close to the middle of the distribution. The median recent state-level figure in this site's data is 2.38% — just above the crossover. Which means the choice between these two structures is not a choice between a safe option and an aggressive one in some obvious sense. It is a close call that resolves differently depending on a variable with a genuine chance of landing on either side.
The downside is asymmetric. At −3%, A loses 26.6% of the capital committed and B loses 106.6% — more than everything invested, because the loan balance does not shrink when the value does. That asymmetry is the real argument for caution, and it does not show up anywhere in the 3% column that most analyses stop at.
One more reading of that table, which is the practical one. The row that should decide the structure is not the 3% row or the 5% row — it is the 0% row, because that is the only one containing no assumption at all. At 0%, A returns 29.1% over a decade and B loses money. Everything above that line is the market's contribution rather than the property's, and an investor who chooses B is choosing to be paid by the market rather than by the building.
That is a legitimate choice. It is a different choice from the one most people think they are making.
The general form: leverage widens the outcome distribution without shifting its centre. It does not make a deal better. It makes the good outcomes better and the bad outcomes worse, and the decision is about which of those you are positioned to absorb.
6. Appreciation is a measurement, not a forecast
Everything in section 3's second table rests on 3%, so it is worth being precise about what that number is.
Published appreciation figures — including the FHFA state-level series behind this site's own dataset — are measurements of what already happened. Applying one forward is a scenario, and it is legitimate to do so as long as it is labelled that way.
The spread is the part that gets lost. Across this site's sourced 50-state data the recent annualised figures run from −2.4% to +7.31%, with a median of 2.38% and a mean of 2.19%. That is not a small dispersion around a national number — it includes states that went backwards.
Three consequences for how to use it:
Run the zero column first. What does the deal return if the price never moves? That is the part of the return that depends on no assumption about the future at all, and it is the honest baseline.
Run a negative column too. Not because prices will fall, but because a structure that survives −3% and one that does not are genuinely different risks, and you cannot tell them apart from the 3% column alone.
Never compound a measurement into a projection and present it as a plan. A ten-year model at 3% is a scenario. The same model described as "what this property will return" is a forecast, and nobody can make one.
7. Not every dollar comes back
One more distinction, because it changes the arithmetic and it is routinely missed.
The down payment converts. It becomes equity, and it returns to you at sale alongside the principal you retired and whatever appreciation occurred.
Acquisition costs are consumed. Closing costs, transfer tax where it is yours, recording tax, and up-front rehab all leave permanently. They raise your cost basis for tax purposes but they never come back as equity.
In the worked comparison, both properties carry $9,000 of that. On Property B's $62,000 of invested capital, $9,000 — roughly 14.5% — never appears in the equity column at all. Fold it into the down payment and you overstate the return by exactly that amount.
This is why a proper hold-period calculation carries acquisition cost as its own line rather than lumping it with the down payment. The five terms — cash flow, principal paydown, appreciation, selling costs, and acquisition costs — should sum exactly to the profit. If they do not, something is hidden in a residual.
8. What this article deliberately leaves out
Two omissions, both intentional.
Tax. Depreciation can turn a cash-positive property into a paper loss; the passive-loss rules decide whether you can use it; recapture takes some of it back at sale at up to 25%. All of that is real and all of it is excluded here, because the figures above are pre-tax and mixing the two produces numbers that are wrong in ways that are hard to see. The tax mechanism is worked through separately.
Rent growth and expense inflation. Both tables hold rent, expenses, and debt service flat across ten years. Escalating them would require a rent-growth rate and an expense-inflation rate, and this site has a sourced figure for neither. Flat understates rent and understates expenses, and the two partially offset — which is a defensible simplification but a simplification nonetheless.
Stating both is not a disclaimer for its own sake. An analysis that quietly included optimistic rent growth would produce much better numbers than these, and it would be the same analysis with an invented input doing the work.
9. How to use all of this
Judge the operating business on cash flow. Cap rate, cash-on-cash, and DSCR describe whether the property works as a business at today's rent and today's rate. A property that loses $362 a month is costing you $4,344 a year regardless of anything else, and that is a real constraint on how many you can own.
Judge the investment on total return. Cash flow plus paydown plus appreciation, across your actual intended hold, with acquisition costs on their own line.
Use the no-appreciation column as the test. It separates deals that work from deals that need the market to cooperate. Both can be reasonable; only one of them is a plan.
Match the structure to the hold. Paydown is worth little over five years and a great deal over twenty. If you know you are a short holder, weight cash flow. If you genuinely intend to hold for decades, the paydown column earns its place.
Be honest about which bet you are making. The comparison in section 3 is not a puzzle with a right answer. It is two different bets, and the version of it that goes wrong is the one where somebody chose B while telling themselves they were choosing A.
Frequently asked questions
Is negative cash flow ever acceptable? It can be a deliberate position, and it should never be an accident. The test is whether you can state the amount per month and the number of years you are choosing to fund, before you buy. If you can, it is a position. If you cannot, it is a hope — and it becomes an urgent problem the first time something breaks.
Why doesn't cap rate include appreciation? Because cap rate describes current income against current value, and appreciation has not happened yet. Building an assumption about the future into a metric used to compare properties would make the comparison meaningless — you would be comparing your optimism, not the buildings.
How much appreciation should I assume? Zero, first. Then your state's published recent figure as a scenario, and then something negative. The range across states in this site's data runs from −2.4% to +7.31%, which should be enough to discourage adopting any single number as a plan.
Is principal paydown really a return? Yes — it is equity, funded by the tenant, and it comes back to you at sale. It is illiquid until then, which is why it feels less real than cash flow, and it accelerates every year, which is why it is worth more on a long hold than most people expect.
Should I put more or less down? That is a question about your capital, your opportunity set, and your view on appreciation, not one an article can answer. What the comparison shows is the shape of the trade: more down improves cash flow and reduces the percentage return in an appreciating market; less down does the reverse and loses money if prices go flat.
Why exclude tax from all of this? Because these are pre-tax figures and combining pre-tax and post-tax numbers in one model produces errors that are difficult to spot. Depreciation, the passive-loss limits, and recapture are all significant and all belong in a separate calculation.
Why hold rent flat for ten years? Because escalating it would require a rent-growth rate this site has no source for, and an invented growth rate applied across a decade does more work in the model than any real input. Flat is conservative on the income side and optimistic on the expense side, and it is honest about both.
What is the single most useful thing here? The no-appreciation column. It takes one extra run and it tells you whether you are buying a business or a bet.
What to do next
The hold-period return calculator runs exactly this analysis on your own numbers — cash flow, principal paydown, and the sale, across three appreciation scenarios side by side, with acquisition costs on a separate line and a flag when the deal only profits because of appreciation.
- Rental analysis calculator — the operating business, which has to work on its own terms first
- Investment cash-to-close — where the acquisition cost that never comes back is established
- Cap rate, cash-on-cash, and DSCR explained — the three metrics this article says are incomplete, and what each is genuinely for
- How rental property is taxed — the layer deliberately excluded here
- How our figures are sourced
This article is general education about rental property returns, not investment, tax, or financial advice, and nothing here is a recommendation about any property, market, or capital structure. Worked figures are arithmetic on illustrative inputs stated on the page, not predictions about any real property. Appreciation figures are measurements of past state-level change from this site's own sourced dataset; applying them forward is a scenario and not a forecast. All figures are pre-tax. Consult qualified professionals about your own situation before buying.