Restaurant Wages by Metro: Cooks, Dishwashers and Servers

This comparison uses the BLS May 2025 Occupational Employment and Wage Statistics tables, released May 15, 2026 and checked October 11, 2026. May 2025 was the latest annual reference period available at that check. The three metros are selected examples, not a ranking of US restaurant labor markets.

Hourly median pay in three metros

Source: the BLS metro bulk download, MSA_M2025_dl.xlsx, hourly median field. The occupation codes are 35-2014 (Cooks, Restaurant), 35-9021 (Dishwashers) and 35-3031 (Waiters and Waitresses). The area's cross-industry estimates cover these occupations wherever they occur in the surveyed economy. They are not restricted to independent restaurants.

Use the complete metro name when comparing a site. The New York-Newark-Jersey City estimate covers a New York–New Jersey labor market, not just New York City's five boroughs. Chicago's area also crosses a state boundary. A metro observation cannot establish the wage floor at a particular address. Check the rules applicable to the actual workplace separately.

Read the spread before choosing a planning rate

The median is the estimated midpoint of worker wages, not the arithmetic average. A 25th-to-75th-percentile range shows the middle of the wage distribution. It is not a pay ladder from beginner to expert, an advertised salary range or a confidence interval around the median. The BLS definitions distinguish a mean from these percentile boundaries.

For a cook opening, compare the actual duties, expected hours and pay structure with the occupation before using its median. A restaurant-cook estimate does not separately price every line station or prep responsibility. The table also cannot tell you what offer a particular applicant would accept, whether a vacancy will fill or what a neighboring restaurant pays today. Use your recent recruiting experience and current comparable openings to test the planning assumption.

Keep server earnings separate from employer cash wages

OEWS measures straight-time gross pay and includes tips. Its wage definition excludes overtime pay, shift differentials and employer benefit costs. The New York server median of $22.89 therefore cannot be copied into a cash-wage budget as though the restaurant necessarily pays the entire amount. Nor should you add an assumed tip amount to $22.89 and call that the BLS estimate. BLS wage concepts.

Keep three amounts visible in your own worksheet: the employer-paid wage, employee tip income and additional employer costs. A job listing that states a base rate plus estimated tips is describing a different measure from base pay alone. For your restaurant budget, reconcile the employer-paid amounts and associated costs to your payroll records. The labor cost percentage guide explains the accounting boundary in more detail.

What would a $2 cook raise do to the weekly budget?

Start with the hours affected, rather than multiplying a metro wage by total headcount. In this hypothetical restaurant, the proposed increase is $2 per hour across 120 straight-time cook hours per week, spread across employees without overtime. That adds $240 in weekly employer-paid wages. Assume, only for this example, that incremental employer costs equal 18% of the added wages.

The result isolates the wage decision with hours and sales held constant. If the new rate changes staffing, overtime, service capacity or retention, those effects need their own evidence. Do not credit an assumed sales lift or lower turnover to the raise before observing it. Pair the payroll view with food cost through prime cost, keeping both costs and sales in the same period.

How to reproduce the wage comparison

1. 1.Open the BLS OEWS tables and choose May 2025, Metropolitan and nonmetropolitan area, XLSX. Use the MSA workbook, not the nonmetropolitan workbook.
2. 2.Select area codes 35620, 31080 and 16980. Retain cross-industry NAICS 000000, ownership 1235 and detailed occupation rows for 35-2014, 35-9021 and 35-3031. Ownership 1235 includes federal, state and local government plus the private sector.
3. 3.Read H_MEDIAN for the first table and H_PCT25, H_MEDIAN and H_PCT75 for the New York detail. Keep hourly dollars to two decimals. Do not substitute H_MEAN, mix national with metro rows or turn an unavailable estimate into zero.
4. 4.Retain the reference month, publication date and full area names when exporting a comparison. Recheck the current BLS release before reusing these figures for a later budget.

The May 2025 technical notes explain that OEWS combines six survey panels over three years and adjusts earlier wages to the reference date. These are modeled survey estimates, not a live census of job offers. The workbook's relative standard error for the mean does not provide a confidence interval for the median. Small differences between metros should not be presented as statistically established pay gaps.

Turn the benchmark into your own labor plan

Write down the role, workplace, chosen rate and weekly paid hours, then record the source and date of each assumption. Add employer costs from your own records and compare the total with sales for the same week. Keep a subtotal for each role, but use the whole restaurant's labor when reporting its overall labor percentage. A kitchen-only subtotal divided by total sales describes only the kitchen portion.