Know these first
- L1.3.02 Simple and compound interest
- L1.3.03 Present value and discounting
Math Atoms
By the end you will be able to
- 1Convert between discrete rates (compounded m times a year) and continuous rates.
- 2Explain continuous compounding as the limit of compounding more and more often, and where e comes from.
- 3Grow and discount money with continuous rates, using e^(r_c T) and e^(−r_c T).
- 4Explain why continuous rates and log returns are convenient: they add up over time.
Why it matters
In 1683 the mathematician Jacob Bernoulli asked a banker's question. If a bank paid 100% a year, and added the interest more and more often, how rich could you get? The answer is not “infinitely rich”. It is 2.71828…, a number now called e.
Core explanation
Compounding without limit
In L1.3.02 we saw that adding interest more often helps, but each step helps less: 6% monthly beats 6% quarterly by only a little. Push that idea to its end. Add interest every day, every hour, every second, every instant. That is continuous compounding.
Bernoulli's version makes the pattern easy to see. Start with 1 at 100% a year:
| Interest added | m | 1 grows to |
|---|---|---|
| Annually | 1 | 2.000000 |
| Semi-annually | 2 | 2.250000 |
| Quarterly | 4 | 2.441406 |
| Monthly | 12 | 2.613035 |
| Weekly | 52 | 2.692597 |
| Daily | 365 | 2.714567 |
| Hourly | 8,760 | 2.718127 |
| Every minute | 525,600 | 2.718279 |
The growth keeps rising but slows to a crawl, closing in on a ceiling of 2.718281…. That ceiling is the number e. At any rate r, compounding without limit for one year multiplies money by . At 5%, that is 1.051271, an effective annual rate of 5.127%.
Why would anyone use it? No bank account literally adds interest every instant. But continuous rates are the standard language of derivatives pricing and risk models, for two reasons. The formulas are cleaner: growth is a single , with no m. And continuous rates add up. Grow at 3% for a year and then 4% for a year, and the total is exactly . Discrete rates do not add like that.
The same idea applies to investment returns. The continuously compounded return, or log return, of a price moving from 100 to 110 is ln(1.10) = 9.53%. If the price then falls to 99, the log return is -10.54%. Add them and you get -1.01%, exactly the log return from 100 to 99. The simple returns, +10% and −10%, add to zero, which is wrong: the investor lost 1%.
Because each convention gives a different number for the same growth, you will often need to convert. Two rates are equivalent if they turn 1 into the same amount over a year.
The formal version
Two new tools appear here: the number and its partner, the natural logarithm ln. The Math Atom on e and continuous growth covers both. We write for a continuous rate and for a rate compounded times a year.
Why continuous rates add. One line of exponent rules:
So growing at then at is the same as growing once at their sum. For log returns, .
The formulas, for reference
With : and .
Try it
Start with Bernoulli's 100% and slide the frequency from annual to every minute: the dot climbs towards the dashed ceiling at e. Then set the rate to 5% and read the table: the same rate in every convention.
Try itThe compounding race
Open the rate converter →1 grows in a year to
2.000000
Horizontal axis: m, the number of times interest is added a year.
100% annually is the same as…
| Compounded annually | 100.000% |
| Compounded semi-annually | 82.843% |
| Compounded quarterly | 75.683% |
| Compounded monthly | 71.356% |
| Continuous | 69.315% |
Continuous rate: 69.315%. Check it with the exponents and logarithms calculator.
Worked example
A bond rate into a model rate
A risk analyst in Singapore has a US Treasury yield of 5% a year, compounded semi-annually (the US Treasury convention). Her pricing model needs continuous rates. Try each step before revealing it.
Converting and using a continuous rate
Step 1 of 5- 1
Set up the conversion
Regulatory lens · looking ahead
No rules apply yet. Market risk models, such as value at risk in later levels, are usually built on log returns, and derivative pricing (Black–Scholes) uses continuous rates throughout.
Sparks
SparkA number found by asking about interest
SparkBlack–Scholes discounts with e
Key takeaways
- Continuous compounding is the limit of adding interest more and more often; 1 at 100% for a year becomes e ≈ 2.71828.
- Grow with FV = PV e^(r_c T) and discount with PV = FV e^(−r_c T).
- Convert with r_c = m ln(1 + r_m/m) and back with r_m = m(e^(r_c/m) − 1).
- A continuous rate is always a little lower than the discrete rate that gives the same growth.
- Continuous rates and log returns add up over time, which is why risk and pricing models use them.
Reaction
Check what stuck
Question 1 · objective 3
£1,000 grows at a continuously compounded rate of 6% a year for 5 years. What is it worth at the end, in £?
Question 2 · objective 1
What continuously compounded rate is equivalent to 8% a year compounded annually? Answer in %, to three decimals.
Question 3 · objective 1
A continuously compounded rate is 5%. What is the equivalent annually compounded rate, in %, to three decimals?
Question 4 · objective 2
Which rate grows money fastest over a year?
Question 5 · objective 3
What is €500 due in 2 years worth today at a continuously compounded rate of 3%, in €?
Glossary
- Continuous compounding
- Interest added at every instant, the limit of compounding more and more often. Growth over T years is e^(r_c T).
- e
- The number 2.71828…, the limit of (1 + 1/m)^m as m grows without end.
- Natural logarithm (ln)
- The natural logarithm: the power you must raise e to, to get a number. It undoes e^x.
- Continuous rate (r_c)
- An annual rate compounded continuously.
- Equivalent rates
- Two rates are equivalent if they grow money by the same amount over the same time.
- Log return
- The continuously compounded return over a period: ln(end value ÷ start value). Log returns over successive periods add up.
Deep Dives & Origins
Deep Dives
- John C. Hull, Options, Futures, and Other Derivatives (Pearson), chapter 4, section on continuous compounding. Intermediate.
- Eli Maor, e: The Story of a Number (Princeton University Press, 1994). Beginner, historical.
Origins
- Hull, Options, Futures, and Other Derivatives (Pearson).
- Maor, e: The Story of a Number (Princeton University Press, 1994).
- Black and Scholes, “The pricing of options and corporate liabilities”, Journal of Political Economy 81(3), 1973.
Watch this lesson
Video version
Prefer to watch? The same lesson as a 12-minute video will sit here, made from this page's script.