Uploaded October 2026 | Updated November 2026, 5 hours ago
If you work the doubling time out with continuous compounding, the "right" answer is a rule of 69.3. So why has everyone used 72 for more than five hundred years?
Check out the channel! @polymathematic
The rule of 72 says that if you divide 72 by an interest rate, you get roughly how many years it takes your money to double. At 8 %, that's 72 ÷ 8 = 9 years. Do the math with continuous compounding, A = Pe^rt, and the constant that falls out is the natural log of 2, about 0.693. That makes it look like the honest version should be a rule of 69.3, and like 72 only won because it's friendlier: it divides evenly by 2, 3, 4, 6, 8, 9 and 12, and 69.3 divides evenly by almost nothing.
That's true, but it isn't the reason. The rule of 72 turns up in 15th-century Italy (Luca Pacioli writes it down in 1494), more than a century before anyone had logarithms, and almost two centuries before anyone had the number e. The merchants and accountants who used it weren't compounding continuously. They compounded once a year, and they could check the rule by brute force: multiply by 1.08, then do it again, and again. After nine years you're at 1.999. Almost exactly double, almost exactly when 72 ÷ 8 says you would be.
So the better question is what the yearly compounding formula says. Start with A = P(1 + r)^t. Doubling means A = 2P, the P's cancel, and you're left with 2 = (1 + r)^t. Take the natural log of both sides and you get 0.693 = t · ln(1 + r), so the doubling time is exactly t = 0.693 / ln(1 + r). That's still not something you'd want to do in your head.
Here's where a Taylor series earns its keep. Near zero, ln(1 + r) = r − r²/2 + r³/3 − …, and for everyday interest rates the first two terms already do most of the work: ln(1 + r) ≈ r(1 − r/2). Divide by that, and approximate 1/(1 − r/2) by 1 + r/2, and the doubling time becomes t ≈ 0.693(1 + r/2) / r.
Look at that numerator. It isn't a fixed number at all; it depends on the rate. When the rate is close to zero, it's 69.3, which is exactly the continuous-compounding answer. Plug in 8 %, though, and you get 0.693 × 1.04, which is about 0.72. So the "72" in the rule of 72 is just the right numerator for rates around 8 %, which is right where a lot of investments live.
The same formula tells you when to swap the numerator. At 2 %, the numerator works out to about 70, so for something like inflation you'd want a rule of 70: at 2 % inflation, prices double in roughly 70 ÷ 2 = 35 years. The Italian accountants found the rule by brute force. Calculus is what tells us why it works.
#ruleof72 #compoundinterest #taylorseries
Watch more Math Videos:
Math Minis: youtube.com/playlist?list=PLrc8spN1b3jkQynJ5heNvSs72gCPj_hwj
Math Minutes: youtube.com/playlist?list=PLrc8spN1b3jmVFYwHiuMzCYJ0y_yiwUge
Number Sense: youtube.com/playlist?list=PLrc8spN1b3jksKkY_oarNFEFAHjrIV97U
MATHCOUNTS: youtube.com/playlist?list=PLrc8spN1b3jlqDkZXtby9lnEaOBy7-ZY0
Follow Tim Ricchuiti:
TikTok: tiktok.com/@polymathematic
Mathstodon: https://mathstodon.xyz/@polymathematic
Instagram: instagram.com/polymathematicnet
Reddit: reddit.com/user/polymath-matic
Facebook: facebook.com/polymathematic
If you work the doubling time out with continuous compounding, the "right" answer is a rule of 69.3. So why has everyone used 72 for more than five hundred years?
Check out the channel! @polymathematic
The rule of 72 says that if you divide 72 by an interest rate, you get roughly how many years it takes your money to double. At 8 %, that's 72 ÷ 8 = 9 years. Do the math with continuous compounding, A = Pe^rt, and the constant that falls out is the natural log of 2, about 0.693. That makes it look like the honest version should be a rule of 69.3, and like 72 only won because it's friendlier: it divides evenly by 2, 3, 4, 6, 8, 9 and 12, and 69.3 divides evenly by almost nothing.
That's true, but it isn't the reason. The rule of 72 turns up in 15th-century Italy (Luca Pacioli writes it down in 1494), more than a century before anyone had logarithms, and almost two centuries before anyone had the number e. The merchants and accountants who used it weren't compounding continuously. They compounded once a year, and they could check the rule by brute force: multiply by 1.08, then do it again, and again. After nine years you're at 1.999. Almost exactly double, almost exactly when 72 ÷ 8 says you would be.
So the better question is what the yearly compounding formula says. Start with A = P(1 + r)^t. Doubling means A = 2P, the P's cancel, and you're left with 2 = (1 + r)^t. Take the natural log of both sides and you get 0.693 = t · ln(1 + r), so the doubling time is exactly t = 0.693 / ln(1 + r). That's still not something you'd want to do in your head.
Here's where a Taylor series earns its keep. Near zero, ln(1 + r) = r − r²/2 + r³/3 − …, and for everyday interest rates the first two terms already do most of the work: ln(1 + r) ≈ r(1 − r/2). Divide by that, and approximate 1/(1 − r/2) by 1 + r/2, and the doubling time becomes t ≈ 0.693(1 + r/2) / r.
Look at that numerator. It isn't a fixed number at all; it depends on the rate. When the rate is close to zero, it's 69.3, which is exactly the continuous-compounding answer. Plug in 8 %, though, and you get 0.693 × 1.04, which is about 0.72. So the "72" in the rule of 72 is just the right numerator for rates around 8 %, which is right where a lot of investments live.
The same formula tells you when to swap the numerator. At 2 %, the numerator works out to about 70, so for something like inflation you'd want a rule of 70: at 2 % inflation, prices double in roughly 70 ÷ 2 = 35 years. The Italian accountants found the rule by brute force. Calculus is what tells us why it works.
#ruleof72 #compoundinterest #taylorseries
Watch more Math Videos:
Math Minis: youtube.com/playlist?list=PLrc8spN1b3jkQynJ5heNvSs72gCPj_hwj
Math Minutes: youtube.com/playlist?list=PLrc8spN1b3jmVFYwHiuMzCYJ0y_yiwUge
Number Sense: youtube.com/playlist?list=PLrc8spN1b3jksKkY_oarNFEFAHjrIV97U
MATHCOUNTS: youtube.com/playlist?list=PLrc8spN1b3jlqDkZXtby9lnEaOBy7-ZY0
Follow Tim Ricchuiti:
TikTok: tiktok.com/@polymathematic
Mathstodon: https://mathstodon.xyz/@polymathematic
Instagram: instagram.com/polymathematicnet
Reddit: reddit.com/user/polymath-matic
Facebook: facebook.com/polymathematic










