Present Value vs. Future Value: Compare, Calculate and Understand the Difference

Present value vs future value at a glance

Present value and future value are the same time-value relationship solved in opposite directions. Future value asks what an amount you hold today becomes at a later date. Present value asks what an amount dated later is worth today. One equation answers both, and which one a question needs depends only on which quantity you already know.

The distinction earns its keep through a rule you can apply before any arithmetic: before comparing two cash flows, move them to the same date. Money at different dates is not comparable, and most bad financial comparisons are an attempt to compare amounts that were never brought to a common date — a lump sum against instalments, a rebate now against a discount later, a pay rise this year against a bonus in three.

Which date you choose does not change the ranking — today is conventional only because the answer arrives in units you can spend. What is never valid is valuing one option at one date and the other at another.

One equation, two directions

Start with one period. An amount PV growing at rate r becomes PV(1 + r). Leave it for a second period and the new balance grows by the same factor. Repeat n times and the factor carries an exponent, and everything else in time-value work is that one factor applied forwards or backwards.

Compounding multiplies by the factor. Discounting divides by it. Because multiplication and division by the same number are exact inverses, these are not two models with two sets of assumptions — they are one model read left to right or right to left.

Three of the four quantities are always known and the fourth is what you solve for. Fix PV, r and n and the answer is FV; fix FV, r and n and it is PV; fix PV, FV and n and you get the implied rate; fix PV, FV and r and you get the periods required. The last two are not new formulas, only the same equation solved for a different letter.

The explorer below runs whichever direction you select and then reports the opposite operation applied to the other end of the horizon. It lands on the same figure to the cent, because both directions divide and multiply by one number.

What is present value?

Present value is what a cash flow dated in the future is worth today, given a rate at which money can be moved through time. It is not a judgement about whether the payment will arrive. It is the size of the deposit that would grow into that payment over the same horizon at the same rate.

That is the cleanest way to read any present value. $46,043.49 is the present value of $75,000 due in ten years at 5% because $46,043.49 invested at 5% for ten years becomes $75,000. The two figures are the same money seen at two dates, and neither is more real than the other.

Present value is lower than the future amount whenever the rate is positive, and the gap is not a fee or a penalty for waiting. It is the return the money would have earned in the meantime, subtracted because you did not have it to invest.

The mechanism is a single number, the discount factor: the reciprocal of the growth factor, and therefore what one dollar due at a given date is worth now. Multiply any future amount by it and the present value falls out.

What is future value?

Future value is the same relationship read the other way: what an amount held today becomes by a stated date, at a stated rate. The input is a plan and the output is a balance.

It is a projection, not a prediction. The formula holds no view about markets, no probability distribution and no uncertainty band; it compounds the assumptions handed to it and nothing else. That narrowness is what makes it useful: because the arithmetic is mechanical, every disagreement about a result is a disagreement about an assumption — the rate, the horizon or the deposit — and can be argued on those terms.

It also means a future value is only as durable as the rate behind it. Over five years an error in the rate is a rounding difference. Over thirty it dominates the answer, because the error compounds along with everything else. $10,000 at 6% for thirty years is $57,434.91; at 7% it is $76,122.55. One point of rate, a third more money.

PV and FV should round-trip

The strongest check available on any time-value calculation costs one extra step: solve it in the other direction and see whether you land back where you started. If discounting the future value at the same rate over the same horizon does not return the original amount, one of the two calculations is wrong.

Take $10,000 at 6% for five years, compounded annually. The growth factor is 1.06^5 = 1.338226, so the future value is $13,382.26. Discount that back at the same 6% over the same five years — divide by 1.338226 — and the result is $10,000.00. The round trip closes to the cent because both directions use one factor.

It closes at every intermediate date too, which is the property worth carrying into a real comparison. The balance at the end of year three is $11,910.16, and its present value at 6% is $10,000. There is no date at which this money is worth something other than $10,000 in today's terms at this rate.

When a round trip fails to close, the cause is almost always one of three things: the rate was expressed per year in one direction and per period in the other, the number of periods differed between the two, or a payment stream was valued with a single-amount formula. All three announce themselves in the size of the mismatch — a factor of ten or more points at the rate, a small persistent gap points at the timing convention.

Which one should you use?

Use future value when the input is a plan and the output is a balance: what a deposit becomes, what a savings schedule reaches, what a portfolio looks like at 65. Use present value when the input is an obligation or an offer and the output is what it is worth now: what a settlement is worth against instalments, what a pension stream is worth as a lump sum, how much a bill years away requires today.

A quicker test is to ask which date the answer has to be stated at. If it is a number you would write on a plan — a target balance, a projected pot — the question is a future value. If it is a number you would hold up against a price, an offer or a balance you have now, it is a present value.

Needing both directions is common and not a contradiction. Whether to accept a buyout is a present value question; whether the buyout leaves you better off at retirement is a future value one. Same inputs, same rate, two different dates — and, as the rule requires, one date per comparison.

Discount rate vs growth rate

The symbol r is identical in both directions and its job is not. In a future value calculation it is a growth assumption: what the money is expected to earn. In a present value calculation it is an opportunity cost: the return given up by not having the money now.

That difference has a practical consequence people find counter-intuitive. Two people can discount the same payment stream at different rates and both be right, because they have different alternatives. Somebody carrying credit-card debt and somebody holding cash for a house deposit genuinely face different costs of waiting, so the same offer is genuinely worth different amounts to them.

The sensitivity is severe at long horizons. $100,000 due in 25 years is worth $47,760.56 today at 3%, but only $23,299.86 at 6% — less than half, from a three-point change in an assumption nobody can observe directly.

This is why a present value quoted without its rate is not a result. The rate is not a detail of the calculation; it is most of the answer.

Real vs nominal

Discounting for opportunity cost and adjusting for inflation are the same operation applied with different rates, which is exactly why they get double counted. Discount a future amount at a nominal return and then deflate the answer for inflation, and you have charged for inflation twice.

The relationship between the two rates is multiplicative, not additive. A 6% nominal return against 2.5% inflation is not a 3.5% real return; it is 1.06 ÷ 1.025 − 1 = 3.4146%. Subtraction is close enough at low rates to hide the error and wrong enough at high ones to matter.

The rule that follows is simple and admits no exceptions. Nominal cash flows are discounted at a nominal rate. Real cash flows — amounts expressed in today's money — are discounted at a real rate. Both frames give the same present value when applied consistently, and mixing them is the only way to get a wrong number.

Worked through: $50,000 of spending, in today's money, needed twenty years from now, with 2.5% inflation and a 6% nominal return. In the nominal frame the bill is 50,000 × 1.025^20 = $81,930.82, discounted at 6% to $25,546.42. In the real frame the bill stays $50,000 and is discounted at 3.4146% — also $25,546.42. The two frames agree exactly, as they must.

Choosing between them is convenience, not correctness. The real frame suits long-horizon personal decisions, because the answer arrives in units you already understand. The nominal frame suits contractually fixed cash flows — a loan payment, a level annuity that will not rise with prices — which are nominal by construction.

Compounding frequency

A rate means nothing until you know how often it is applied. “6% compounded monthly” means 0.5% applied twelve times a year, and twelve applications of 0.5% is not 6% — it is 6.1678%. That figure is the effective annual rate, and it is the only basis on which two differently quoted rates can be compared.

Two conversions do all the work. The rate per period is the nominal annual quote divided by the compounding periods in a year, and the number of periods is years multiplied by that same figure. Take both from one frequency and the calculation is consistent; take the rate from one and the period count from another and it is not.

The common version of that error is an annual rate with monthly periods — 6% and 240 rather than 0.5% and 240 — which overstates a future value by orders of magnitude while still looking plausible in a spreadsheet cell.

One further distinction is worth holding. A rate quoted as a nominal annual rate is divided by the frequency; a rate quoted as an effective annual rate is already the annual figure, so its periodic rate is (1 + EAR)^(1/m) − 1 instead. On a six-year monthly savings plan at 5%, reading the quote as nominal gives a required deposit of $596.91 and reading it as effective gives $598.97 — $2.06 a month, or $148.32 more paid in across the 72 deposits, from nothing but a convention.

Frequency matters less at short horizons than people expect and more at long ones. Over five years the difference between annual and daily compounding of 6% on $10,000 is $116.00. Over thirty years the same gap is $3,052.61.

Single cash flow vs payment stream

Most real decisions involve a series rather than one payment: a mortgage, a pension, a rent, a spending plan in retirement. Valuing one is the same operation repeated — each payment discounted from its own date — and for a level series that sum collapses into a closed form.

The closed form matters because a twenty-five-year monthly stream is 300 separate discountings, and the annuity factor does all of them in one multiplication. It is why a loan balance and an investment projection run on identical machinery: a lender quoting a payment has solved this equation for PMT given the amount borrowed, and testing whether an offer is fair solves the same equation for PV given the payment.

Six cash-flow shapes cover almost everything you will meet. The first four are the working set. The last two are limiting cases, useful mainly as sanity checks on the first four.

A perpetuity is a level payment with no end date, and its present value is simply PMT ÷ i — the annuity factor as N goes to infinity. A growing perpetuity adds a constant growth rate, PMT ÷ (i − g), and is defined only while i exceeds g. They are worth knowing because they bound the annuity: $1,500 a month at 5% compounded monthly is worth $360,000 as a perpetuity against $227,287.97 over twenty years, so 63% of the value of a payment that never stops arrives in the first two decades. It also shows how badly a naive total misleads — $360,000 is what you get by adding twenty years of payments up, and it is what the same payment would be worth if it ran forever.

Ordinary annuity vs annuity due

An ordinary annuity pays at the end of each period; an annuity due pays at the start. That single period of difference applies to every payment in the series, so the whole result scales by (1 + i) — one factor applied once, not one extra payment.

Discounting $1,500 a month for twenty years at 5% nominal compounded monthly gives $227,287.97 as an ordinary annuity and $228,235.00 as an annuity due. The $947.03 difference is exactly one month of interest on the ordinary value. Compounding $500 a month for ten years at 6% gives $81,939.67 against $82,349.37, a $409.70 gap on the same logic.

Getting the convention wrong is a small error in percentage terms and a stubborn one: it never washes out, and it is always signed the same way. Rent, insurance premiums, leases and most subscriptions are annuities due. Loan payments, salaries, bond coupons and most investment contributions are ordinary annuities. When a contract does not say, the timing of the first payment settles it — a first payment today means an annuity due.

Present value vs NPV

Present value is the value of a cash flow, or a set of cash flows, at a chosen date. Net present value is the same discounting applied to a complete cash-flow set — inflows and outflows together — so that the answer comes out net rather than gross.

The word doing the work is net: a present value says what something is worth, a net present value what it is worth after the cost of getting it — provided that cost is inside the set being summed. Under the usual convention the initial cost sits at t = 0, carries a negative sign, and is discounted by (1 + r)^0 = 1, which is to say it enters at face value.

Worked through: an outlay of $90,000 today, returning $25,000, $30,000, $35,000 and $40,000 at the end of the next four years. At 8% the four inflows have a present value of $106,053.64; net of the $90,000 outlay the NPV is $16,053.64. At 10% it is $11,137.22 and at 12% it is $6,570.28 — identical cash flows, separated by nothing but the rate.

Two conventions are in circulation and both are legitimate: the initial cost sits inside the summation at t = 0, as above, or the gross present value of the inflows is stated and the cost subtracted outside the sum. The arithmetic is identical as long as the cost is counted exactly once, at its own date. The failure mode is subtracting an outlay that is already inside the summation, which charges it twice and can flip the sign of the answer.

A positive NPV means the cash flows are worth more than they cost at the rate used. It does not mean the decision is a good one, and the sign itself can turn on the rate — which is why an NPV is quoted with its discount rate, for the same reason a present value is.

Worked examples

Five calculations, each stated with its inputs, its factor and its check. Every figure here was produced by the same calculation modules the site's calculators run on, and every one round-trips.

One — a lump-sum future value. $25,000 at 5% compounded annually for twelve years. The growth factor is 1.05^12 = 1.795856, so the future value is $44,896.41. Divide that by the same factor and $25,000 comes back exactly.

Two — the reverse, a present value. What is needed today to have $75,000 in ten years at 5% compounded annually? The factor is 1.05^10 = 1.628895, so the present value is 75,000 ÷ 1.628895 = $46,043.49. Compounded forward for ten years that becomes $74,999.99 — a cent short only because the present value itself was rounded to the cent.

Three — a savings goal, which is the annuity factor solved for the payment. Reaching $50,000 in six years at 5% nominal compounded monthly, depositing at the end of each month: the rate per period is 0.4166667% over 72 periods, the future-value annuity factor is 83.764259, and the required deposit is 50,000 ÷ 83.764259 = $596.91. Seventy-two deposits total $42,977.52, so $7,022.48 of the target is growth. Feeding $596.91 back through the schedule reaches $49,999.72, twenty-eight cents short purely because the deposit is rounded to the cent.

Four — an annuity valued in both directions. $500 a month for ten years at 6% nominal compounded monthly: the rate per period is 0.5% over 120 periods. The future-value factor is 163.879347, giving $81,939.67 at the end. The present-value factor is 90.073453, giving $45,036.73 today. They describe one stream, and $45,036.73 × 1.005^120 = $81,939.67 confirms it. Deposits total $60,000.

Five — a lump sum against a payment stream, which is the comparison this whole page is built for. $250,000 today, or $1,500 a month for twenty years. The stream totals $360,000, and the larger number looks obviously better until the payments are moved to a common date. At 5% nominal compounded monthly the rate per period is 0.4166667% over 240 payments; the present-value factor is 151.525313, so the stream is worth $227,287.97 and the lump sum wins by $22,712.03.

Change the discount rate to 3% and the same stream is worth $270,466.37, and the answer reverses. Nothing about the offer changed — only the assumption about what the money could otherwise earn. The break-even sits between 3% and 4%: at 4% the stream is worth $247,532.79 and the lump sum wins by $2,467.21, a margin small enough that the rate assumption, not the offer, is deciding the question.

The choice of valuation date changes the units but never the ranking. Compounded forward to year twenty instead of discounted to today, the lump sum is worth $678,160.07 and the stream $616,550.50 — a gap of $61,609.57, which is the $22,712.03 gap multiplied by the same twenty-year growth factor. Any date works, as long as it is one date.

Choosing a discount rate you can defend

Because the discount rate drives the answer, it deserves a stated justification rather than a default. The question it answers is narrow: what could this money reliably earn if you had it today, in the use you would actually put it to?

For someone with expensive debt the honest rate is close to the debt rate, because repaying it is the guaranteed alternative use. For someone holding cash against a near-term obligation it is a deposit rate, and for a long-horizon investor with no immediate use for the money it is closer to a portfolio return.

Matching the risk of the rate to the risk of the cash flow is the step most often skipped. A guaranteed payment stream discounted at an equity return will look worse than it is, because you have implicitly charged it for risk it does not carry. The reverse error is rarer and worse: discounting an uncertain cash flow at a risk-free rate prices it as though the payment were certain.

Two habits make a rate defensible. State it as a sentence — “the alternative use of this money is repaying a 6.5% loan” — rather than as a bare number, so the justification travels with the figure. Then show the result at a rate above and below it. If the decision survives the range, the rate was not load-bearing. If it flips inside the range, the honest conclusion is that the comparison does not settle the question, and something other than arithmetic will have to.

Common mistakes

Every mistake below produces a number that looks reasonable. That is what makes them expensive: none of them fails loudly, and most survive a second look.

The first is the one the others are variations of, and the operational rule disposes of it: before comparing two cash flows, move them to the same date.

Edge cases

The closed forms behave sensibly at their boundaries, but several of those boundaries have arithmetic of their own and one of them divides by zero.

At a zero rate the growth factor is 1 and present value equals future value: money at different dates really is interchangeable. The annuity factor, however, is (1 − (1 + i)^−N) ÷ i, which is 0 ÷ 0 at i = 0. Its limit is N, so a level stream is worth PMT × N. A spreadsheet handles this for you; a hand-rolled formula returns an error unless it special-cases the zero.

Negative rates are arithmetically ordinary and intuitively backwards. At −1%, $10,000 held for five years falls to $9,509.90, while $10,000 due in five years is worth $10,515.36 today — more than the amount itself, because holding money now costs you. Nothing in the formulas changes; only the reading does.

At n = 0 the factor is (1 + r)^0 = 1 and present value equals future value at any rate. That is not a curiosity: it is precisely the rule that lets an initial cost dated t = 0 enter an NPV at face value.

Fractional periods are well defined by the same exponent. Five and a half years at 6% is a factor of 1.06^5.5 = 1.377788, giving $13,777.88 on $10,000. Note that this is not the midpoint of the year-five and year-six balances, which is $13,783.73: compounding is convex, so interpolating linearly between two annual figures always overstates the value in between.

At very long horizons the discount factor does nearly all the work and the rate assumption becomes the answer: $1,000,000 in a hundred years is worth 45 times as much at 3% as at 7%. A present value at that horizon should be read as an illustration of the rate, not as a valuation of the cash flow.

Check the result in Excel or Google Sheets

Both directions ship with Excel and Google Sheets under the names FV and PV, taking the same arguments in each: rate, nper, pmt, then the opposite value and the timing switch. Two of those arguments cause nearly all the trouble.

rate and nper have to come from the same period: for a monthly calculation, rate is the annual rate divided by twelve and nper is a number of months. Neither program warns you when they disagree.

type is the timing switch, and it is the last argument. type=0, the default, means payments at the end of each period — an ordinary annuity. type=1 means payments at the start — an annuity due. All type changes is a single multiplication by (1 + rate), which is why leaving it out gives an answer wrong by exactly one period of interest rather than an obviously broken one.

The sign convention is the other trap. Both functions treat money paid out as negative and money received as positive, so =PV(0.05/12, 240, 1500) returns −$227,287.97. Entering the payment as −1500, as below, returns the positive figure: a result right in magnitude and wrong in sign is a convention issue, not an arithmetic one.

One more asymmetry is worth memorising: NPV() discounts its first value by one full period, so it assumes the range starts at t = 1. A cost dated t = 0 therefore belongs outside the function, added to its result, which is why the NPV formula below has a trailing + C1.

Frequently asked questions

Eight questions that come up repeatedly, answered against the figures used above rather than in the abstract.