Points vs cash back: what the same spend actually returns

The comparison on one page: the same $20,000 of spend

Every row below spends the same $20,000. Only the earning style and the exit door change. The formula for every points row is the same: dollars spent, times points earned per dollar, times the value each point returns when redeemed. Cash back skips the third term, which is both its weakness and its strength.

Read the table as a range, not a verdict. The two flight rows reuse the worked examples in our how many points you need for a flight worksheet: $412 cash against 28,000 points and $56 in fees for the 1.27 cent row, and $2,850 cash against 85,000 points and $210 in fees for the 3.11 cent row. A points household does not live in one row; it lives in the row its next real booking lands in. If your bookings usually land in the 1.27 cent row, your honest comparison is $254 against $400, and cash back is winning your wallet today whatever the premium cabin row promises.

The break even line: what a point must be worth

Divide the cash back percentage by your earn rate and you get the value each point must return just to tie. At 1 point per dollar against a 2 percent cash back card, the break even value is a demanding 2 cents per point, a figure the 1.27 cent economy example on this page falls well short of. At 2 points per dollar, the line drops to 1 cent, which published travel rates already meet. Some portal rates sit above that, and the guide to pricing a Chase Travel fare against a transfer shows how a boosted rate turns into cents per point for one flight. This is why the points case usually stands or falls on earn rate first: category bonuses and flat 2x earning do the heavy lifting before any transfer magic is asked to help.

Use your own cash back alternative in the formula, not the 2 percent used here. If your realistic cash back card pays 1.5 percent, the line at 1x is 1.5 cents, and a great deal of ordinary award travel clears it. If it pays more in your top categories, the line rises there. The comparison is always between the two cards you would actually hold, on the spend you actually have. The printed Amazon and gift card rates for five banks give the dollar floor for the comparison.

The tie line, drawn against the floors the issuers print

The break even table can be drawn, and drawing it against the issuers' own floors shows where the argument ends before any booking is priced. Each curve below is a cash back percentage divided by an earn rate. Each dashed line is a rate an issuer prints for a plain exit: Chase says Ultimate Rewards points are typically worth 1 cent each when redeemed for cash back, and American Express prints 0.6 cents per point for covering eligible charges on a statement. Where a curve has dropped below a dashed line, that floor alone ties or beats the cash back card at that earn rate, and no booking has to do any work.

The 2 percent curve meets Chase's 1 cent line at exactly 2 points per dollar, and Amex's 0.6 cent line at about 3.33 points per dollar (2 divided by 0.6). Read across: spend earning 2 Chase points per dollar ties a 2 percent card even if every point leaves as cash, before any annual fee. Spend earning 2 Amex points per dollar returns only 1.2 percent at the statement credit rate, so its points need exits worth 1 cent or more, such as the 1 cent Amex prints for flights booked through Amex Travel, just to tie. The same earn rate, two different verdicts, because the floors differ.

Two currencies are missing from the chart on purpose. Capital One's Venture page lists statement credits among its redemptions without printing a rate, and Bilt's help center says it does not publish an exact single value for a Bilt Point. For those, the floor is whatever your own redemption screen shows on the day, and the Cash or Points Calculator takes that number as an input rather than assuming one.

Add the annual fee: the break even formula with a fee in it

The table above ignores fees. Real comparisons rarely can. Put the fee into the break even value directly: break even cents per point = (cash back earned on the same spend + the points card's annual fee) / points earned. The fee raises the line, and the more you spend, the less each dollar of fee moves it.

Arithmetic illustration with stated inputs, not a card recommendation: a card earning a flat 2 points per dollar with a $95 annual fee, against a 2 percent cash back card with no fee. On $20,000 of spend, cash back pays $400 and the points card earns 40,000 points, so the points must return (400 + 95) / 40,000 = 1.24 cents each to tie. On $40,000 of spend, the line falls to (800 + 95) / 80,000 = 1.12 cents. The same fee costs a light spender more than a heavy one, which is the arithmetic behind the common advice to check the fee against your own spend, not a typical household's.

A whole budget, category by category

Labelled illustration with stated inputs. A household spends $30,000 a year: $8,000 on groceries, $4,000 on dining, and $18,000 on everything else. Card A pays a flat 2 percent cash back with no fee: $600. Card B, a hypothetical points card, earns 4 points per dollar on groceries and dining and 1 point per dollar elsewhere, with an annual fee of $250 as a stated input. Card B earns 32,000 + 16,000 + 18,000 = 66,000 points.

The break even value is (600 + 250) / 66,000 = 1.29 cents per point. Below that, cash back wins; above it, points win. Redeemed at the 0.6 cent statement credit floor, the points return $396 minus the $250 fee, or $146. At a 1 cent floor, they return $660 minus $250, or $410, still $190 behind cash back. At 1.5 cents on bookings the household really takes, they return $990 minus $250, or $740, ahead by $140. The household's choice therefore turns on one honest question: how many of its points have actually left at 1.29 cents or better over the last year? The spend optimizer on this site runs the same category split with recorded earn rates.

The floors the issuers themselves publish

A transferable point is not worth an average. It is worth at least its published cash floor, and more only when a specific booking pays more. Chase says points are typically worth 1 cent each for cash back, and that Points Boost can make them worth up to 50 percent more on select Chase Travel bookings with eligible cards. American Express publishes a ladder: 0.6 cents per point for covering charges, 0.7 cents for prepaid hotels and other purchases through its portal, 0.7 to 1 cent for Pay with Points at checkout, 1 cent for flights through Amex Travel, and up to 1 cent for gift cards. The Amex Travel against a partner award for one fare uses that 1 cent flight rate as the bar a transfer has to clear. Citi's Double Cash page pays its 1 percent purchase cash back as 1 ThankYou point per dollar and gives the example of 10,000 ThankYou Points for a $100 statement credit, which works out to 1 cent per point, although Citi's card page says redemption rates may vary. Capital One's Venture page says miles can be redeemed for statement credits, gift cards, travel and more, or transferred to 15+ travel loyalty programs, without printing a cash rate, and its redemption guide adds covering an eligible travel purchase within 90 days; read the rate for each inside your own account.

These floors discipline the whole comparison. If your realistic exit is the floor, compare floors with cash percentages directly and the argument is over in one line. Everything above the floor is booking dependent upside: real, sometimes large, and never guaranteed at the moment you choose which card to put spend on.

Cash back and points are both called rebates in everyday speech, but the tax documents draw finer lines. A 2010 IRS private letter ruling treated a card rebate of usually 1 percent of purchases as an adjustment to the purchase price, and the Tax Court in Shankar included in income an airline ticket redeemed with Citibank points that it assumed were a premium for a bank account. The quoted texts, and what each one leaves open, are in are points and miles taxable.

When transferable points win

Points win first through earn rate. Category multipliers change the multiplication before redemption is even discussed: American Express publishes 4 points per dollar for its Gold Card at restaurants worldwide, on up to $50,000 of purchases per calendar year, and at U.S. supermarkets, on up to $25,000 per calendar year, with 1 point per dollar after each cap. Four points redeemed at a plain 1 cent return 4 percent, double a flat cash back rate, and even at the 0.6 cent statement credit floor they return 2.4 percent. Spend that concentrates in bonus categories is where a points card builds its lead, up to the cap.

Points win second through bookings whose award price ignores the cash fare: the premium cabin example in the table returns 3.11 cents per point because the award price moved less than the fare. They win third through transfer bonuses applied to a booking that already exists, where a bonus lowers the effective points cost of a chosen redemption. Notice what all three wins require: spend in the right categories, a trip you are actually taking, or a booking already found. The upside is earned at redemption time, by work, which is exactly what the cash back percentage never asks of you.

When cash back wins

Cash back wins whenever the points exit is weak or distant. A balance redeemed at a 0.6 or 1 cent floor, earned at 1 point per dollar, returns 0.6 to 1 percent, half or less of a flat 2 percent card, and no future intention changes that arithmetic. It wins when travel is rare, because the upside lives entirely in bookings. It wins when a balance sits: points can be devalued by the program at any time and can be lost when the account holding them closes, depending on the issuer's rules, while cash already redeemed cannot be devalued by anyone.

Cash back also wins on friction. It needs no award search, no transfer timing, no partner account, and no redemption decision, and it spends anywhere, including on the rent, the repair, or the fare sale that points cannot improve. Add an annual fee to the points side and the gap widens mechanically: a $95 fee on $20,000 of spend costs 0.48 percentage points of return. Citi's own comparison guide frames the choice the same way: the right card depends on your spending habits and how you want to use your rewards, with cash back cards described as possibly the more straightforward option.

Edge cases where the simple comparison breaks

Category caps cut the lead. A 4 point category with a $25,000 annual cap earns 4 points per dollar only up to the cap. A household spending $30,000 a year in that category earns 100,000 points on the first $25,000 and 5,000 on the rest, an effective 3.5 points per dollar, not 4. Check the cap and the rate after it before running the multiplication on a large category.

Spending outside the bonus categories decides more than people expect. In the budget example, $18,000 of the $30,000 earns only 1 point per dollar on Card B, and that portion returns 1 percent at a 1 cent floor, against 2 percent on the cash back card. A points card's average return is pulled toward its base rate by every dollar that misses a category, which is why the category split, not the headline multiplier, is the honest input.

Redemption friction is the last edge case. A points balance only reaches its booking value if you search, compare, and transfer at the right moment, and points kept for a later trip stay exposed to devaluation and to account closure, as the points expiration guide explains. Cash back has no such exposure once redeemed, though a returned purchase takes back cash back and points alike; the return clauses of Amex, Chase and Bilt show how the deduction works and how a balance can go negative. If your redemptions are irregular, value your points at the floor you actually use, not at the booking you hope to make. When a real booking does appear, the how many points you need for a flight worksheet and the Cash or Points Calculator price it against its cash fare.

Score your points by how they actually left

The 1.29 cent break even value from the budget example is only useful against a number you can defend, and the defensible number is a weighted average: the cash your points actually replaced, divided by the points that left. One strong booking and a pile of cash-outs do not average to the strong booking. The table is an arithmetic illustration with stated inputs, using Card B's 66,000 points a year from the budget example and the two flight examples from the top of this page.

In the one-year path the weighted value is 1.12 cents, under the 1.29 line. Card B's points replace $736 of spending, minus the $250 fee is $486, and the $600 from the cash back card wins by $114. In the two-year path the household holds two years of points, 132,000, and one block becomes the premium seat. The weighted value rises to 2.41 cents: $3,186 replaced, minus $500 in fees, is $2,686 against $1,200 of cash back over the same two years, so points lead by $1,486. Same card, same spend; the verdict turns on whether the premium trip gets booked.

That is why the weighted average belongs in the decision and the single best redemption does not. It also prices the cost of waiting: the two-year path needed a balance held at the bank for a year or more, exposed to the program and ratio changes described in save your points or use them now. If the seat had not been there when the time came, those 85,000 points would have left at the floor, and the two-year total would look like two copies of the first year.

Four ways a points-versus-cash verdict goes wrong

The classic mistake is comparing a points card's best category against a cash back card's base rate, then declaring a winner for all spend. Run both cards across your whole budget, category by category, the way our spend optimizer does with recorded earn rates. The second mistake is valuing every point at your best ever redemption. One 3 cent booking does not make a balance that mostly leaves at 1 cent a 3 cent currency; weight the value by how your points actually leave.

The third mistake is counting a welcome offer as an annual return. A bonus is a one time payment for opening an account, available on cash back cards too, and it says nothing about which earning style wins year two. The fourth is forgetting that the two styles blend: some cash back is literally paid in transferable points, as the Citi example shows, so the real decision is often not points or cash but which doors you want open for the same earnings. Decide with the break even table, your last year of redemptions, and your next year of trips, in that order.

Before you decide which way your spend should earn

• Write your annual spend and the earn rate it actually receives, category by category, not the headline rate.
• Multiply earn rate by a redemption value you have evidence for, starting with your currency's published cash floor.
• Compare the result with a flat cash back percentage on the same spend, fees included on both sides.
• Subtract any annual fee as a percentage of your spend before declaring a winner.
• Choose points only if a booking you would really take clears the break even value in the table.