How it works
What Horalis Growth does with your numbers, what it assumes, and where every figure comes from.
The answer first
My money starts with the form. The result opens with the answer: what you could have, what it grows by and what it gives you a month.
“Today, from what you have” uses the rate in “What it can pay you”, on today's money, with no years.
Then the years as a table, and the related questions: the worst, typical and best period of the past, and “What if…?”.
Your money grows
You start with an amount and add some every month. Each year the money grows by a percentage, and that growth grows too.
All amounts are in today's money, in the currency you choose. Prices rise over time, so the growth shown is growth after rising prices.
Each monthly amount goes in at the end of its month, and it rises with prices.
How much it grows
Nothing is chosen for you: tap an example or type your own growth a year, after rising prices.
Each example shows its worst year in the data beside its growth: the two always go together.
An example uses its own past years. Any other number moves up and down like the S&P 500.
For reference, world stocks grew 5.2% a year from 1900 to 2024 (UBS Global Investment Returns Yearbook 2025). Type 5 and the page names it.
Beside the result: your amounts in the past as it happened, started in every year of the data. You see the worst, the typical and the best period.
The years start at 20: every example has 16 or more such runs in its data. Change them to your plan's.
Each choice grows at its average from past years. All use the same years, 1988–2022, so none looks better for starting in a good decade.
Stocks go up and down. So Horalis Growth works out 1,000 possible futures, each shuffling past years.
“If it goes badly” is where 1 in 10 of those futures end below. “If it goes well” is where 1 in 10 end above.
“First 10 years like 2000–2009” replays that decade as it happened, then grows at the average. Bonds and gold did well then, so they take their worst ten years.
A mix grows like its parts, by weight. A savings account earns its interest, minus rising prices. Your own numbers can replace any of these.
Only kinds of assets with a long public history are shown: the S&P 500 (US stocks), German government bonds, gold and savings. Never a single company or fund.
What it can pay you
Each month you could take out part of your money: your money × the rate ÷ 12.
The rate starts at the most that lasted 30 years from every start in the investment's past, the worst included (Bengen, 1994). Few starts in the data make it a rough guide, and the page says so. The past is not a promise.
Why 30 years: a long retirement, as in Bengen (1994) and the Trinity study (1998).
Your own growth has no history, so the rate is yours. The page shows what a fall like the worst year of the most similar asset would do at the start.
The slider goes from 2% to 7%, plus the data's rate.
“It lasted 30 years” says in how many of 100 possible futures that amount lasted 30 years.
Prudent: in 90 or more. Risky: in 75 or more. Very risky: in fewer.
Your goals
Under the result, My goals: only the ones you add. Each says when your plan gets there.
You give the price of what you want. The grey example only shows the kind of figure.
Living in a country uses what an average person there lives on (see Countries). Living without working uses your own monthly costs, or a country's as a start.
Nothing is taken from your money: each goal is worked out on its own.
Check your plan
Under the result, up to three observations, only when they apply. Each has its amounts. They say what is, never what to do.
Few years: your plan's years, or a goal, under 5 years away, and money that goes up and down. It counts how many of the 1,000 possible futures end below what you put in by then. It shows from 1 in 10. The method.
Savings for 10 years or more: what the account keeps in today's money, against what you put in. And against the S&P 500, with the same amounts: what it gives up, and their worst fall. That fall comes from yearly data: within a year it may have been deeper. The method.
Countries
The table has 122 countries: those the World Bank's official data can price. No figure is invented for the others.
Each cost is what an average person there lives on a month, housing included. It comes from the household survey mean (2010–2025, each country its latest).
That mean is brought to 2024 prices with the World Bank's price levels.
It is a national average: cities differ a lot. Some countries measure spending, others income, and a mean is pulled up by those who have most.
✓ and a year mean what your money pays each month covers it from that year on. Otherwise the table says when your plan gets there. The method.
Rising prices in each country
Each country has a yearly rise in prices for the long run: the goal its central bank publishes, checked in September 2026.
Euro countries use the European Central Bank's 2%. Where there is no goal, it is about the 2015–2024 average.
In Angola, Burundi, Ethiopia, Ghana, Haiti, Iran, Liberia, Malawi, Nigeria, Pakistan, São Tomé & Príncipe, Sierra Leone, Türkiye, Ukraine, Uzbekistan, and Zambia, prices rose 10% a year or more in 2015–2024. Where a goal exists, the goal is still used.
Test my plan
Your plan goes through past years in the order they happened, with no replay. Figures are for whole calendar years.
Each crash starts the year before it. The fall is what the crash did to the money you had at the top.
A fall that came back within one year does not show in yearly figures. Covid in 2020 is one of them.
What it does not do
It gives no advice and makes no forecast. It leaves out taxes and most fees.
It downloads nothing while you use it: every figure is part of the page, with its source and year.
Amounts are in the currency you choose. Growth is each investment's own, after rising prices: changes between currencies are not counted.
Sources
Each source, what it gives, how far its data goes, and its terms.
UBS Global Investment Returns Yearbook 2025 (Dimson, Marsh and Staunton), as reported by Cambridge Judge Business School
World stocks' yearly growth after rising prices, 1900–2024: 5.2%
Data: 1900 to 2024. Terms: Only this one figure is used: the page names it when you type 5%.
Robert J. Shiller, Yale University
S&P 500 yearly growth (Shiller's S&P Composite), and US prices until 2022
Data: to June 2023. Terms: Published openly on his site; used with credit.
OECD and Deutsche Bundesbank
German 10-year government bond yields
Data: to December 2024. Terms: OECD data: CC BY 4.0, free to use with credit.
Destatis, the German statistics office
German prices
Data: to December 2024. Terms: Data licence Germany, attribution 2.0: free to use with credit.
US Bureau of Labor Statistics
US prices for 2023 and 2024
Data: to December 2024. Terms: Public domain.
World Bank, Commodity Price Data (the Pink Sheet)
Gold price, December average of each year
Data: to December 2024, read from a public copy of the World Bank's data. Terms: CC BY 4.0: free to use with credit.
Open the source: World Bank, Commodity Price Data (the Pink Sheet)
World Bank, Poverty and Inequality Platform and World Development Indicators
What an average person lives on (household surveys), price levels (ICP 2021) and each year's rise in prices
Data: surveys to 2025, prices 2024. Provisional: read from a public copy of the World Bank's data, until the yearly download. Terms: CC BY 4.0: free to use with credit.
Open the source: World Bank, Poverty and Inequality Platform and World Development Indicators
Central banks
Their targets for rising prices
Data: checked September 2026. Terms: Public information.
Unicode CLDR
Country names in each language
Data: as built into Node.js. Terms: Unicode License: free to use with credit.
Data licenses
Most sources let anyone reuse their data with credit: the World Bank, the OECD, Destatis, the US Bureau of Labor Statistics and Unicode.
Robert Shiller publishes his data openly on his site; it is used with credit. From UBS, only one published figure is quoted: 5.2%.
Horalis Growth always says where a figure comes from and of which year.
The method and the sources are public. The code belongs to Horalis. The data keeps its owners' rights.