Every Sharpe Ratio You've Ever Seen Picked a 'Risk-Free Rate.' In Indonesia, Which Number Is That Actually Supposed to Be?
5 min read
Loading...
5 min read
Your portfolio made 10% this year. Your friend says its Sharpe ratio is 0.81. A blog says it is 0.53. An app says 0.37. Same portfolio, three answers.
Nobody made a maths error. Each one used a different risk-free rate, and in Indonesia there are at least three numbers people use for it.
The risk-free rate is the return you could earn taking no real risk. The Sharpe ratio subtracts it from your return before dividing by volatility (how much your portfolio bounces around). Change that one input and the score moves.
In theory it is simple. It is the return on an investment where you know exactly what you will get, and when.
In practice, three tests help.
It must be safe for you, not just for the bank. The loss risk must be close to zero for the person using the number.
It must match your time horizon. If you measure a one-year Sharpe ratio, the rate should be for about one year. A 10-year bond can lose value if you sell it after one year, so it is not risk-free over that window.
You must be able to actually earn it. A benchmark you cannot buy is only a reference, not an alternative.
Keep these three tests in mind. Now the candidates.
Bank Indonesia (BI, the central bank) held its policy rate at 5.75% at its 22 to 23 September 2026 meeting, according to BI's announcement as reported by Warta Kepri. Its deposit facility rate, the rate BI pays banks for parking cash overnight, stayed at 4.75%.
It looks like a natural anchor. It is safe, it is short term and it is published by the central bank. Our own Sharpe ratio explainer uses it for exactly that reason.
The catch is the third test. You cannot deposit money at the BI Rate. It is a policy rate for banks, not a product for savers. It is a good reference, but you cannot earn it.
SBN stands for Surat Berharga Negara, Indonesia's government bonds. A government that borrows in its own currency is the closest thing to "safe" in rupiah.
The 10-year SBN yield was about 7.08% on 25 September 2026, according to Trading Economics and Bareksa's market report the same day. Other outlets that week described it as stable near 7%.
This number is real and you can buy it, through retail bonds. But it fails the horizon test for a one-year Sharpe ratio. A 10-year bond pays more partly because it carries price risk: if rates rise, its market price falls. That is a risk, so it is not "risk-free" over one year.
It also says something about how a yield works. Read more in our guide to the US 10-year Treasury yield as the global risk-free bar.
This is the one most Indonesians actually use for their own savings.
Deposits are covered by LPS, the deposit insurer, up to Rp2 billion per customer per bank. LPS also sets a guarantee rate: the highest rate at which a deposit is still fully insured. It was 3.50% for rupiah deposits at commercial banks for 1 June to 30 September 2026, according to Kontan's report on the LPS decision.
Be careful here. That 3.50% is a ceiling for insurance, not the average rate banks pay. Check the actual rate on the deposit you would really open. LPS resets the guarantee rate every few months, so the number after 30 September may differ.
A deposit passes the "safe" and "can earn it" tests well. Tenors are flexible, so you can match your horizon too.
Here is a made-up portfolio. It returned 10% over the year, with 8% volatility.
(10% − Risk-free rate) ÷ 8%
| Risk-free rate used | Rate | Return above it | Sharpe ratio |
|---|---|---|---|
| Bank deposit guarantee rate | 3.50% | 6.50% | 0.81 |
| BI Rate | 5.75% | 4.25% | 0.53 |
| 10-year SBN yield | 7.08% | 2.92% | 0.37 |
The portfolio did not change. The verdict went from "good" (above 1.0 is solid, 0.5 to 1.0 acceptable) to "borderline" just by swapping one input.
This portfolio and its numbers are an illustration, not real results. The three rates are the sourced figures above.
There is no single right answer. But there is a wrong habit: picking whichever number makes your result look best.
A practical rule:
For a one-year Sharpe ratio, use a short-term rate you can really earn. For most people that is a time deposit or a short-tenor government instrument.
For long horizons, use a government yield of similar maturity. That is the fair bar for a long-term investor.
Use the BI Rate as a reference, not as a promise. It is transparent and easy to track.
Whatever you pick, use the same one every time and write it down. A Sharpe ratio is only useful for comparing your portfolios against each other over time, and that needs a fixed yardstick.
Also treat any Sharpe ratio you read online with one question: which risk-free rate did it use? If the source does not say, the number is hard to compare with yours. The same holds for related scores like the Sortino ratio, which uses the same input.
Rates change. The BI Rate, SBN yields and deposit rates all move through the year, and each one moves your Sharpe ratio with it.
Open NetWort's macro page to see the current BI Rate and government yields side by side. Then pick one as your yardstick and keep it.