The advice to save 20% before buying is repeated so often that it rarely gets tested. It is a real question with a numerical answer, and the answer turns on one thing most people never put a number on.
Short version: on a typical loan, PMI costs about 2.85% of the purchase price in total. If you wait three years to reach 20% and the house rises by more than 0.94% a year while you save, waiting cost you more than the PMI would have — before counting a single month of rent.
PMI feels like a penalty, and penalties feel like something to avoid at any cost. But it is a finite, calculable amount, and once you know the amount you can compare it against what avoiding it costs.
Take a $350,000 house with 10% down. The loan is $315,000, the rate is 6.5%, and PMI at 0.40% a year adds $105 a month. Send the cancellation request the month you reach 80% of the original value and the premium runs for 95 months.
Total PMI paid: $9,975.
That is the entire cost of the thing being avoided. It is not nothing. It is also 2.85% of the purchase price — roughly what a house might move in a single moderate year.
Here is the comparison that decides it. If you wait to save the full 20%, the house does not wait with you. Whatever it gains while you are saving is a cost you pay in the purchase price, and you pay it with interest for thirty years.
So: how fast does the house have to rise before waiting costs more than the PMI?
| If you wait | Waiting is the more expensive choice above |
|---|---|
| 2 years | 1.41% a year |
| 3 years | 0.94% a year |
| 4 years | 0.71% a year |
These are not forecasts. They are thresholds — the rate at which the two paths cost the same. Above the threshold, waiting loses. Below it, waiting wins.
What makes the threshold so low is that PMI is temporary and a purchase price is permanent. You carry $9,975 of premium for eight years; you carry the price of the house for thirty. A small permanent number beats a larger temporary one more often than intuition suggests.
Nobody can tell you what your local market will do. What the table tells you is how little it has to do for the waiting strategy to fail.
There is a second problem with waiting, and it is the one that catches people out. Twenty percent is a percentage, not an amount. As the price rises, so does the deposit you need.
| Growth | Price in 3 yrs | 20% deposit | Extra |
|---|---|---|---|
| 0% a year | $350,000 | $70,000 | — |
| 1% a year | $360,605 | $72,121 | $2,121 |
| 3% a year | $382,454 | $76,491 | $6,491 |
| 5% a year | $405,169 | $81,034 | $11,034 |
At 3% a year you spend three years saving toward a number that has moved $6,491 further away. At 5% the target runs faster than many households can save. This is the mechanism behind the feeling that the deposit is never quite enough — in a rising market it genuinely is not.
And none of this counts rent. Every month spent waiting is a housing payment that builds nothing. Whatever your rent is, multiply it by the waiting period and add it to the cost of waiting. For most people it is the largest number on this page and it settles the question by itself.
Price the PMI on your own loan
Enter your numbers and see all four cancellation routes with the date you qualify for each.
Buy now and wait for 20% are not the only two choices. There is a structure that avoids PMI without waiting: split the borrowing in two. A first mortgage at 80% of the price, a second mortgage covering the next 10%, and your 10% deposit. It is usually called 80/10/10, or a piggyback.
Because the first mortgage never exceeds 80%, no mortgage insurance is required. The trade is that the second mortgage carries a higher rate — often around two points above the first, and frequently variable.
On the same $350,000 house, with the second at 8.5%:
| Monthly | Cost by month 95 | Cost over 30 years | |
|---|---|---|---|
| 10% down with PMI | $2,096 | $478,866 | $726,740 |
| 80/10/10 piggyback | $2,039 | $474,498 | $734,008 |
The order reverses. The piggyback is cheaper by $57 a month and stays cheaper for about eight years — $4,367 ahead by the time the PMI would have ended. Then it loses, and by year thirty it is $7,268 behind.
The reason is structural rather than incidental. PMI stops. A second mortgage does not. Once the premium falls off the first path, the piggyback carries on paying above-market interest on that second loan for another twenty-two years.
Which makes the piggyback a genuinely good choice if you expect to sell or refinance within roughly eight years, and a poor one if you are buying the house you mean to stay in. That is a question about your life rather than about the arithmetic — but the arithmetic tells you which answer to give it.
Two cautions. Second mortgages are often variable-rate, so the comparison above assumes a stability the product may not have. And a second lien can complicate a later PMI cancellation request on the conventional path, which is worth knowing before you take one on deliberately.
The arithmetic leans one way, but it is not the whole decision. Waiting genuinely wins when:
What does not belong on that list is the idea that PMI is money thrown away and therefore always worth avoiding. It is worth avoiding at the right price. The point of the numbers above is to tell you what that price is.
The premium is not a fixed sentence. How long you carry it is largely under your control, and on this loan the gap between the slowest and fastest routes is more than $6,000.
The short version: do not wait for the automatic cutoff at 78%, request cancellation at 80%, and send extra principal if you can, because it pulls the cancellation date forward by years. The full set of exits is in how to remove PMI, the cost side is in how much does PMI cost, and if the house has appreciated since you bought, cancelling with a new appraisal may be faster than either.
One thing to check before treating an FHA loan as the way around a thin deposit: FHA mortgage insurance mostly does not come off at all. That comparison is in FHA MIP vs. conventional PMI.