Inflation Calculator: What It Costs Later, What It's Worth Now

An inflation calculator applies compounding to prices. You give it an amount, an annual rate, and a number of years, and it tells you what that amount becomes -- either the larger number you would need later to buy the same thing, or the smaller number a future sum is worth in today's money.

The two directions are the same equation. One multiplies by a growth factor, the other divides by it. The reason both appear is that people ask the question both ways: "what will this cost in fifteen years" and "what is a $50,000 payout in 2041 actually worth to me now".

The rate you enter is an assumption you control, not a forecast. Published price indexes measure what already happened, and the past average of any index is not a promise about the next twenty years. Testing a range of rates is more informative than finding the single correct one, because there isn't one.

#The formula

Formula: Future cost = A x (1 + r)^n and Present value = A / (1 + r)^n

Where:

• A is the amount you entered.
• r is the annual inflation rate as a decimal, so 3% is 0.03.
• n is the number of years.
• (1 + r)^n is the cumulative price factor over the period.

Four more outputs come straight off that factor:

• Cumulative price change: (1 + r)^n - 1, expressed as a percent.
• What a dollar today is worth then: 1 / (1 + r)^n.
• Cumulative loss of buying power: 1 - 1 / (1 + r)^n.
• Years for prices to double: ln(2) / ln(1 + r).

The structure is identical to the growth math in the compound interest calculator. The difference is what is compounding: there, a balance you own; here, a price you have to pay.

#A worked example done by hand

Take $1,000, an assumed 3.0% annual inflation rate, and 20 years.

1. Growth factor: 1.03^20. Working it in stages, 1.03^10 = 1.343916, and squaring that gives 1.343916^2 = 1.806111.
2. Future cost: 1,000 x 1.806111 = $1,806.11.
3. Cumulative price change: 1.806111 - 1 = 0.806111, or 80.6%.
4. What a dollar is worth then: 1 / 1.806111 = 0.5537, so about 55.4 cents.
5. Loss of buying power: 1 - 0.5537 = 0.4463, or 44.6%.
6. The other direction: $1,000 received 20 years from now is worth 1,000 / 1.806111 = $553.68 in today's money.
7. Doubling time: ln(2) / ln(1.03) = 0.693147 / 0.029559 = 23.4 years.

Every one of those figures comes from the single number 1.806111. If you can compute the growth factor, you can produce the whole output by hand.

#Prices rising and buying power falling are not the same percentage

This is the most common misreading of the output. Prices rose 80.6% in the example, but buying power fell 44.6%. Those describe the same event from opposite ends, and they are not the same number.

The reason is that the two percentages have different denominators. A price rise is measured against the old price. A loss of buying power is measured against the old purchasing power, which is now the reciprocal.

Worked example: Prices rise 50%. An item that cost $100 now costs $150. Your $100 buys 100 / 150 = 0.6667 of that item, so you have lost 1 - 0.6667 = 33.3% of your buying power, not 50%. A 25% price rise is a 20% loss. A 100% price rise is a 50% loss.

The conversion in general is loss = price change / (1 + price change). Check it: 0.50 / 1.50 = 0.3333. It also means a loss of buying power can never reach 100%, no matter how high prices go, while the price change has no ceiling.

Each cell reads as price change first, loss of buying power second.

#Where each input comes from

Amount. Whatever you are testing: a price, a salary, a lump sum you expect to receive, a savings target. The math does not care which.

Annual rate. There is no single correct value, so pick a basis and label it. Many people use the long-run average of a published consumer price index as a central case and then run one lower and one higher scenario around it. The Bureau of Labor Statistics publishes the Consumer Price Index and its component series; that is where the historical record lives. Whatever you choose, it is an assumption.

Years. The horizon you actually care about. Sensitivity to this input is nonlinear: doubling the years far more than doubles the cumulative effect.

Direction. Choose "what it will cost later" when you have a present amount and want a future one. Choose "what a future amount is worth today" when you are discounting a number that already sits in the future -- a pension figure, a payout, a projected balance.

The output is most sensitive to the product of rate and years. One percentage point on the rate over five years is minor; the same point over thirty years is large, as the table above shows.

#How to read the result

The adjusted amount is a same-purchasing-power equivalent, not a prediction of any specific price. It says: if a basket rises at this rate, this is the number that buys the equivalent basket then.

The "still equivalent" figure in the opposite direction is the one to use for offers and projections. A retirement projection quoted in nominal dollars looks impressive largely because of the price factor baked into it; dividing by (1 + r)^n puts it back into units you understand.

#How fast prices double

Doubling time depends only on the rate, not the amount. The Rule of 72 -- years to double is roughly 72 divided by the percentage rate -- is a good shortcut, and it is slightly conservative at low rates.

#Nominal returns, real returns, and why subtraction is wrong

A return quoted in dollars is nominal. What it buys is real. The relationship between them is a ratio, not a difference:

Formula: 1 + real = (1 + nominal) / (1 + inflation), so real = (nominal - inflation) / (1 + inflation)

Subtracting inflation from the nominal return is an approximation. It is close at low rates and drifts as rates rise.

The per-year gap looks negligible, and over one year it is. Compounded, it is not. Take $10,000 at an assumed 5% nominal with 3% inflation over 20 years: the exact real value is $14,690.67, while using a 2% "real rate" from subtraction gives $14,859.47. The shortcut overstates the outcome by $168.81 on a $10,000 starting amount.

The same logic applies whenever you plan in real terms, including any goal you set in the savings goal calculator: either project in nominal dollars and deflate at the end, or project at the exact real rate. Mixing the two is what produces inconsistent answers.

#Why your personal inflation rate differs from a national index

A national consumer price index is an average across a basket weighted to represent an average household. You are not that household.

• Weights. If housing is 40% of your budget rather than the index weight, housing price moves hit you disproportionately.
• Geography. Regional and metro-level price changes diverge from the national figure, sometimes substantially.
• Life stage. Someone paying tuition, someone paying a fixed mortgage, and someone paying market rent face different exposures to the same index.
• Substitution. Indexes account for buyers switching between goods; your own substitution behavior may differ.
• Big-ticket timing. A car or a roof purchased in a particular year applies that year's price change to a large share of your spending at once.

The practical move is to weight the component series you can find against your own spending. If you track what you spend, the exercise is arithmetic. The real hourly wage calculator covers a related idea: headline numbers and lived numbers diverge for structural reasons.

#Sanity-checking a salary increase

A raise is a nominal number. Whether it is an increase in what you can buy depends on the price change over the same period. Run both through the ratio.

For a $62,000 salary, comparing a one-year raise against an assumed one-year inflation rate:

A raise that matches the price change holds you level. Anything below it is a real cut expressed as a positive number. Note also that a raise moves your gross pay, while what you can spend moves with net pay -- run the change through the take-home pay calculator before deciding what it is worth.

#What this model leaves out

• A constant rate. Prices do not rise smoothly. A single average rate reproduces the endpoint of a period, not the path through it, and the path matters if you are spending along the way.
• Any forecast. The model has no view on what inflation will be. It compounds the number you supply.
• Quality change. A product that improves is not the same product ten years later. Index producers adjust for this; a flat rate applied to one item does not.
• Category divergence. Different categories move at very different rates over long periods. One economy-wide number cannot represent all of them.
• Taxes. Nominal gains can be taxed even when the real gain is zero or negative, which lowers the real after-tax outcome further.
• Deflation and negative rates. The formula handles a negative r mathematically, but the economic behavior around falling prices is not something a compounding factor describes.

#Common mistakes

Reading the price change as the loss of buying power. They are different numbers. Convert with loss = change / (1 + change).

Subtracting inflation from a return. Use the ratio, especially over long horizons or at higher rates.

Comparing a nominal projection with a real target. Pick one basis and stay in it.

Assuming the national index is your index. Your basket has your weights.

Treating a past average as a forecast. It describes what happened, not what will.

Applying an economy-wide rate to one specific item. Tuition, insurance, and electricity do not track a general index. For something like household energy, a metered electricity cost calculator using your own rate is more accurate than inflating last year's bill.