Does raising the minimum wage cause inflation? The pass-through evidence

Search "does raising the minimum wage cause inflation" and you get two kinds of answers. One says yes: every wage hike pushes up menu prices, and higher prices are inflation. The other says no: the effect on the overall price level is too small to measure, so the worry is a myth. Both answers dodge the interesting question. A minimum wage increase does raise prices. The research on that is unusually consistent. What it does not do is set off sustained, economy-wide inflation. The useful questions are how much prices rise, where, and who ends up paying.

This piece walks through the theory, collects the best pass-through estimates in one table, and then turns the question around. Since July 2009, inflation has done to the $7.25 federal minimum what no Congress has voted to do: it has cut its real value by about a third. That fact belongs in an honest version of this debate, and so does the dispute over whether to tie the minimum to prices automatically.

Higher prices are not the same as inflation

Economists reserve "inflation" for a sustained rise in the general price level. Milton Friedman's famous line, that inflation is "always and everywhere a monetary phenomenon," is about that kind of rise: prices across the economy climbing year after year because money and spending grow faster than output. A minimum wage increase is a different animal. It raises the cost of one input, low-wage labor, in the industries that use a lot of it. Prices in those industries go up, roughly once, and then settle at a new level.

That is a relative price change: burgers and groceries get more expensive compared with other things. It shows up in the Consumer Price Index as a small, one-time bump, not a new trend. Low-wage industries are a modest share of consumer spending, and most workers already earn more than the minimum, so the bump in the overall index is small. Saying "the minimum wage doesn't cause inflation" is therefore technically right. Saying "the minimum wage doesn't raise prices" is wrong.

We think the first framing lets supporters off too easily. A price rise that is too small to move the national CPI can still be large for the specific purchases of low-income households, and it is still a transfer that someone pays for.

Where the money for a raise comes from

When the law requires a business to pay more for the same hour of work, the extra cost has to come from somewhere. There are four main places:

• Customers, through higher prices.
• Owners, through lower profits, or closures and fewer openings in the long run.
• Workers, through fewer hours, fewer hires, thinner benefits, or more automation.
• Productivity, if the raise cuts turnover or prompts better management.

"Pass-through" is the share that lands on customers. A rough rule of thumb follows from simple accounting. If minimum-wage labor is a fixed share of a business's costs and the business passes all of the increase on, the price rise equals that cost share times the effective wage increase. Here is a purely illustrative example: if the affected wages make up a fifth of costs and those wages rise 10 percent, full pass-through means prices rise about 2 percent. That is why pass-through estimates look small in percentage terms even when firms pass on all of the cost.

Pass-through also tells you something about jobs. In a competitive labor market, higher costs push prices up and sales down, and lower sales mean fewer hours worked. In the monopsony story, where employers hold wages below the competitive level, a minimum wage can raise employment and output, and prices need not rise at all. Aaronson, French and MacDonald made exactly this argument in 2008: rising prices are evidence against the monopsony explanation for small measured job losses. More recent Berkeley work reads partial pass-through as consistent with monopsony. The price data are a battleground for the employment debate, not a side issue.

The pass-through evidence

The table below collects estimates we could verify against each study's own published abstract or text. Where a study reports an effect "per 10 percent" increase in the minimum wage, we give that. Where it reports a share of costs passed on, we give that.

Three things stand out. First, nearly every serious study finds that prices go up. The disagreement is over how much of the cost reaches customers, not whether any does. Second, the effects are biggest where minimum-wage labor matters most: limited-service restaurants rather than sit-down ones, and low-wage areas rather than high-wage ones. Third, the California estimates have moved around. The first Berkeley version, released in September 2024, was quoted at the time as finding a 3.7 percent increase. Later revisions put it at 2.1 and then 1.5 percent. The NBER study, using official CPI data and a broader category that includes restaurants the law doesn't cover, finds more than twice the Berkeley team's latest figure. Our California scorecard goes through each study.

Who pays the higher prices

This is where the "it's only a few cents" defense runs into trouble. Thomas MaCurdy's 2015 paper in the Journal of Political Economy granted supporters their best case: assume no job losses and assume consumers absorb the cost through prices. Under those assumptions, he found that a national minimum wage increase works like a value-added tax on consumer goods, one more regressive than a typical state sales tax. The extra earnings, meanwhile, were spread nearly evenly across the income distribution, because many minimum-wage earners are not in poor households. His conclusion was that this pattern undercuts the case for the minimum wage as an antipoverty tool.

Harasztosi and Lindner's Hungarian evidence points the same way. If roughly three-quarters of the cost is paid by consumers, the policy is mainly a transfer from shoppers to some workers. Fast food and discount groceries are not luxury goods. Customers there are disproportionately the same low- and middle-income households a wage floor is meant to help.

None of this means the price effects are large in absolute terms. A few cents on a burger is a few cents. The point is that the cost is real and badly targeted. A tax-funded wage supplement such as the Earned Income Tax Credit spreads the cost across all taxpayers in proportion to what they pay, rather than charging it at the drive-through.

The other direction: what inflation does to $7.25

The federal minimum rose to $7.25 on July 24, 2009, and it has not changed since. Prices have. Using the official BLS Consumer Price Index for All Urban Consumers (CPI-U, U.S. city average, all items, not seasonally adjusted), here is the arithmetic.

Supporters of a higher minimum present this as a scandal. There is another way to read it. The federal floor has stopped binding for almost everyone. In 2024, according to the BLS, 82,000 hourly workers earned exactly $7.25 and 760,000 reported less, together just 1.0 percent of hourly paid workers, against 13.4 percent in 1979. Inflation has carried out a slow, partial repeal that Congress never voted on. The labor market did not collapse. Wages at the bottom rose because employers had to compete for workers, and the states that wanted a higher floor set one.

The same arithmetic explains a figure you will hear often: the 1968 minimum of $1.60 was "worth more" than today's. At the 1968 annual-average CPI-U of 34.8, $1.60 equals about $14.80 in 2025 dollars. That comparison is real, but incomplete. In 1968 the $1.60 rate applied to workers in jobs covered by the original 1938 law and the 1961 amendments, while workers covered later had a lower rate. Comparing peaks also says nothing about how many people the floor kept out of work, which is the question that matters most to us.

Calculator: convert a wage between years

Should the minimum wage be indexed to inflation?

If erosion is the problem, automatic indexing looks like the fix. Many states already do it. Arizona, Colorado, Connecticut, Oregon and the District of Columbia are among the jurisdictions that adjust their floors every year. The federal Raise the Wage Act of 2019 would have gone further and indexed the federal minimum to median wages once it reached $15. Supporters make reasonable arguments: indexing ends the long freezes followed by big catch-up jumps, gives employers predictable increases, and takes a recurring fight out of politics.

It is still a design choice, not a technicality, and there are good reasons to resist it:

• It locks in the starting level. Indexing a floor set too high doesn't fix the mistake. It makes the mistake permanent in real terms, and it removes the slow correction that inflation has applied to the federal rate since 2009.
• It ignores local conditions. One national price index applied to every labor market preserves a gap that already varies widely. A floor that bites lightly in a high-wage metro can bite hard in a low-wage rural county.
• It mandates raises in bad years. After a price shock, an indexed floor rises automatically, even when the businesses that pay it are squeezed by the same costs. That is when a fixed floor gives employers room to adjust.
• The index itself is a choice. Prices and median wages move differently. Linking to wages can raise the floor faster than living costs. Linking to prices can make it lag behind wages.

Voters and legislators disagree on this too. Missouri voters approved a $15 minimum with annual inflation adjustments in 2024. In 2025, the legislature and Governor Mike Kehoe repealed the indexing provision, keeping the $15 rate. Honest people land on both sides. Our view is that indexing is the right answer to the wrong question. If the goal is the purchasing power of low-income families, the tools that target that directly, such as the EITC, are already indexed to inflation, and they don't price anyone out of a first job.

What it means for the argument

A careful reading of the evidence supports three claims. First, minimum wage increases raise prices in the sectors that employ low-wage workers. Estimates range from partial to full pass-through, and that range is itself evidence in the monopsony debate. Second, that is not inflation in the macroeconomic sense. It is a one-time, sector-specific price increase that shows up only faintly in the national CPI. Anyone warning that a wage hike will set off an inflationary spiral is overstating the case. Third, the cost that does reach customers is regressive in the way MaCurdy describes. It is collected at the register from people who shop where prices are lowest.

Running the other way, inflation has taken roughly a third of the purchasing power of $7.25 since 2009, and the floor now binds for about one hourly worker in a hundred. That is a reason to stop treating the federal minimum as the main lever for low-wage workers' incomes, not a reason to index it. Try your own assumptions in the simulator, or browse the underlying studies in our research tracker.