Job postings are down in tech hubs and up in Boise, Wichita and Reno

Hiring is leaving the tech coasts

The clearest shift in this month's data is where postings are coming back and where they are not. Indeed's postings index, which we set beside our own count, has Seattle-Tacoma-Bellevue 32.4% below its February 2020 level and San Francisco-Oakland-Fremont 28.6% below.

At the other end, Jackson, Mississippi is 36.3% above its February 2020 level, Boise City 36%, Wichita 34.5% and Reno 32.3%.

Postings are 32.4% below their February 2020 level in Seattle and 36.3% above it in Jackson, Mississippi.

This is a comparison with a pre-pandemic baseline, not a monthly trend. It is our first edition, so we have no month-over-month data and will not imply a direction of travel we cannot show. The table below carries each metro's change over the past year as well.

• Seattle-Tacoma-Bellevue: 32.4% below February 2020, but 5.6% above a year ago.
• San Francisco-Oakland-Fremont: 28.6% below February 2020 and 14.9% above a year ago, so the recent direction is up.
• Santa Rosa, Boulder and Portland: 25.8%, 22% and 19% below February 2020. Portland-Vancouver-Hillsboro is also 1.1% below a year ago.
• Boise City and Reno: 36% and 32.3% above February 2020, and 13% and 15.7% above a year ago.
• Jackson, Mississippi: 36.3% above February 2020 and 0.4% below a year ago, so its gain is older than the last twelve months.

What our own sample can and cannot say

Our reading covers 7,678 employers whose career pages we read and 335,906 open postings, 85,182 of them in the 100 largest cities. It is not every job in the country. The Indeed comparison is a separate dataset that we use for context.

A smaller market is not an empty one. San Francisco still shows 17,941 open postings in our sample, more than New York's 12,642. Where it differs from the cities now growing fastest is in how long postings stay open and who they are open to, which the next section covers. See the full metro list for every area.

Coastal postings linger and entry-level doors narrow

Fewer postings than in 2020 is only part of the coastal picture. The postings that remain stay open longer. Nationally, 23.5% of the postings we read have been open more than 60 days, and the median posting has been open 29 days.

In San Francisco, 47% are past 60 days and the median is 56 days. Austin comes in at 43.5% and 50 days, and Seattle at 41.8% and 44 days. It is not only a coastal habit, though: Harrisburg-Carlisle, Pennsylvania leads the list at 49.4% with a median of 56 days.

Long-open postings can mean hard-to-fill senior roles, or listings kept up as a standing call for résumés. Our data cannot tell us which. See the full ranking.

• Quickest to fill: Louisville/Jefferson County has just 12.6% of postings open past 60 days, and Port St. Lucie 12.7%.
• Fastest median: Grand Rapids-Wyoming-Kentwood has a median of 17 days open, and Corpus Christi 15 days.
• Entry level, national: 10.2% of postings we read are entry level.
• Entry level, narrowest: San Francisco and San Jose are both at 6.2%, and Seattle is at 7.2%.
• Entry level, widest: Corpus Christi is at 23%, Port St. Lucie 22.5% and Baton Rouge 22.4%.

Put together, the metros with fewer postings than in 2020 also tend to show slower filling and few openings for people starting out. That is a description of the postings, not of the people applying to them. The rankings for entry-level openings and posting age have the rest of the metros.

What a paycheck really buys

A wage means little without prices beside it. We divide the official BLS median wage for all jobs by the BEA regional price level, where the US average is 100, and call the result paycheck power. It is expressed in average-US dollars.

San Jose-Sunnyvale-Santa Clara tops the 99 metro areas at $76,116. Its official median wage is $84,050 and its price level is 110.4. Boulder follows at $66,139, then Seattle at $65,336, Washington at $65,262 and San Francisco at $64,232.

The bottom is more surprising. Miami-Fort Lauderdale-West Palm Beach is last at $42,609, with a median wage of $48,640 and a price level of 114.2. El Paso comes next at $43,943, followed by Riverside-San Bernardino-Ontario at $45,471, Cape Coral-Fort Myers at $45,706 and Orlando-Kissimmee-Sanford at $45,978.

• El Paso: the lowest median wage in the table, $39,510, but prices at 89.9 lift it above Miami.
• San Francisco: a price level of 115.6, the highest of any metro, and housing at 194.7, yet it still lands fifth.
• San Jose: housing at 211.9 against a US average of 100, and a wage high enough to stay first anyway.
• Cheap is not enough: Springfield, Missouri (88.6), Wichita (88.9) and Jackson, Mississippi (89.0) have the lowest overall prices, but none makes the top five.
• Florida: Miami, Cape Coral and Orlando all sit in the bottom five, with wages between $46,630 and $48,640.

The chart below plots wage against price level. Up and to the left, a paycheck goes further.

This measure uses one price index for everyone and an all-jobs median, not our postings. It will not describe any single household. It does show that where you live can matter as much as what the job pays. The paycheck power ranking and the housing cost ranking hold the full detail.

Where employers still keep pay to themselves

Across everything we read, 37.8% of postings list a pay range. That national figure hides a wide split between places. In San Francisco-Oakland-Fremont, 73.7% of postings list pay, in New York 65.6%, in Denver 63.7%, in Los Angeles 62.4% and in San Jose 62.2%.

At the other end, the share is 3.4% in Lakeland-Winter Haven, 4.2% in Cape Coral-Fort Myers, 4.7% in Fort Wayne, 5.7% in Jackson, Mississippi and 6.6% in Birmingham. Several of those metros have only a few hundred postings in our sample, so we would not put much weight on small gaps among them. The scale of the gap between top and bottom is not in doubt.

• State leaders: the District of Columbia lists pay in 69.3% of postings, California 62.9%, New York 62.1% and Colorado 59.3%.
• State laggards: Mississippi lists pay in 4.9% of postings, North Dakota 7.4%, Alaska 7.7% and Indiana 8.6%.
• Sample sizes: Lakeland has 156 open postings in our sample, Jackson 121 and Fort Wayne 177.
• Prices and pay listing: across metro areas, the correlation between the share listing pay and the local price level is 0.71, a strong link.

The cost of hidden pay falls on anyone trying to compare offers. Where few postings list a range, nobody can easily tell what a job is worth before spending time on it. The pay-listing ranking shows all 99 metro areas, and the state comparison covers the states.

The map below shows each metro as a dot, with gold dots listing pay more often.

Do pay laws change what employers do?

Thirteen states now require pay in postings. The newest is Maine, in force since July 29, 2026. In the 100 largest cities, 65.7% of the 52,591 postings in those states list pay, against 24% of the 32,645 postings in the other states.

That gap is large, but it could simply reflect who lives where. Expensive states are more likely to have passed these laws, and employers there may have been more open about pay anyway. So we ran a tougher test.

The same-employer test

We looked at 329 employers with enough postings in both kinds of state. Together they list pay in 59.5% of their postings in law states (11,956 postings) and 24.7% in other states (11,552 postings).

The typical employer lists pay only 4 points more often in a state with a law than in one without.

That 4-point figure is the median gap across employers. It is far smaller than the pooled difference between 59.5% and 24.7%.

• More in law states: 142 employers list pay more often where a law applies.
• About the same: 171 employers list pay at similar rates in both kinds of state.
• Less in law states: 16 employers list pay less often where a law applies.
• Median gap: 4 percentage points for the typical employer.

One reading is that the laws change behavior at a minority of employers, and sharply, while the rest already do the same thing everywhere. We cannot prove that from this data. Either way, the laws appear to do something, since only 16 of 329 employers go the other way. The table below lists the states at the top and bottom, with the year each requirement began.

Advertised pay and official pay tell different stories

Software developers post the highest median pay of any job family we measure, $202,500, across 7,507 postings with pay. Data scientists follow at $190,000 (538 postings), security and IT at $188,400 (2,284), software engineering at $187,225 (2,866), and legal and compliance and product and design both at $185,000 (1,784 and 3,371).

These are medians of posted pay, not of what people are paid. The sample leans toward software, whose employers account for 24.6% of our postings, and healthcare, at 20.7%. That pulls national averages up.

We compared posted pay for software developers with the official BLS median in the 19 metro areas where we have enough of both. In 17 of them, posted pay is higher.

• Pittsburgh: posted pay is 76.3% above the official median.
• Detroit: 71.3% above.
• Chicago and Austin: 42.9% and 38.5% above.
• Philadelphia: 4.5% below the official median, the only large shortfall.
• San Jose: 1% below, in the one metro where official wages are already very high.

Three things could explain the gap. Employers who list pay may differ from those who do not. Posted ranges may be what an employer hopes to offer rather than what it pays. Official medians include every developer, from junior to senior, while a posting may be aimed at a narrower slice.

The same caution applies to metro-level posted pay, where San Francisco leads at $205,000 and San Jose is at $201,000. El Paso and Pittsburgh both show $180,000, and El Paso's figure is the one we would check first, since it sits beside an official median wage of just $39,510. Adjusted for prices, El Paso's posted pay is $200,196, which tells us more about how few employers list pay there than about the local economy. See the software pay ranking and the posted pay ranking.

Remote work follows expensive housing

Nationally, 23.6% of the postings we read are remote. By metro, San Francisco-Oakland-Fremont has the highest share at 21.5%, followed by New York at 16.2%, Boulder at 15.2%, Winston-Salem at 13.4% and Ann Arbor at 12.4%. The metro figures are lower than the national one, and we would not compare the two directly, since they may be built on different bases.

The pattern across metros is that remote postings go with higher pay and costlier housing. The correlation between remote share and the official median wage is 0.55, and with the housing price level it is 0.42. Those are moderate links, not rules.

• Winston-Salem, North Carolina: 13.4% remote with a housing price level of 71, a clear exception to the pattern.
• Ann Arbor, Michigan: 12.4% remote, housing price level 125.
• Portland, Maine: 10.4% remote, housing price level 105.
• Florida's Gulf and Atlantic coasts: Cape Coral-Fort Myers, North Port-Bradenton-Sarasota and Port St. Lucie all show 0% remote postings.
• Boise City: also 0% remote, even though its postings are 36% above February 2020.

That last pair matters for the lead story. The cities where postings are growing fastest are not growing because of remote work. They are adding jobs that employers want filled in person. See the remote jobs ranking.

The chart below shows the highest ten metros by remote share.

The national backdrop: openings and unemployment

Our postings sit beside official data, and the official data show a steady market. The Bureau of Labor Statistics' JOLTS survey put national job openings in August 2026 at 7,079 thousand, a rate of 4.3%. A year earlier the figures were 6,919 thousand and 4.2%.

Metro unemployment tells a more varied story. The median of the metro areas we publish was 4.1% in August 2026. The rate is 6.5% in Stockton-Lodi, California, and 2.7% in Portland-South Portland, Maine.

• Stockton-Lodi: 6.5%, unchanged from the month before.
• Riverside-San Bernardino-Ontario: 5.9%, up from 5.7%.
• Detroit-Warren-Dearborn: 5.7%, down from 6%.
• Tucson: 5.7%, up from 5.6%.
• Madison, Wisconsin and Savannah, Georgia: both 2.9%. Savannah is also 26.4% above its February 2020 postings level.

We hold back any monthly reading that jumps more than 1.2 points from the month before, as a likely preliminary-estimate blip. Our ranking covers 98 metro areas for that reason.

Two patterns stand out. Savannah pairs very low unemployment with strong postings growth, as you would expect. Detroit pairs high unemployment with posted software pay 71.3% above the official median, which is a reminder that posting data and unemployment data measure different things. The unemployment ranking has every area.

The chart below shows the extremes.

Our read

This is a first edition, so these are judgments from a single snapshot, not forecasts built on a trend. Each one says what would change our mind.

• Hiring keeps spreading toward cheaper metros, but the gap narrows. We expect Boise (36% above February 2020), Wichita (34.5%) and Reno (32.3%) to stay ahead of Seattle (32.4% below) over the next year. But San Francisco is already 14.9% above a year ago and Boulder 10.4%, so we would watch whether the coastal metros close part of the gap. We would change our view if Boise's 13% and Reno's 15.7% yearly gains stall in coming editions.
• Pay listing will keep climbing, unevenly. With Maine now in force as the 13th law state, and 65.7% of postings listing pay in those states against 24% elsewhere, we expect the national 37.8% to drift up. We would not expect big jumps, because the typical employer in our same-employer test lists pay only 4 points more often in law states. We would change our mind if the 142 employers that list more in law states grow in number, or if the 16 that list less do.
• The software pay premium will shrink as more employers list pay. The 17 of 19 metros with posted pay above official wages, led by Pittsburgh at 76.3%, partly reflect which employers choose to list. As listing spreads in low-listing places such as Birmingham (6.6%) and Fort Wayne (4.7%), we expect posted medians to move closer to official ones. We would change our view if premiums stay high in metros where listing is already common.
• The national market drifts rather than turns. Openings at 4.3% against 4.2% a year earlier, and 7,079 thousand against 6,919 thousand, describe a market that is neither cooling quickly nor heating up. We would watch posting age, with 23.5% of postings open more than 60 days, and unemployment in Stockton (6.5%) and Riverside (5.9%). A fall in the openings rate from 4.3%, alongside a rising share of stale postings, would make us more cautious.

The methodology page explains how we read postings and match them to official data, and the interactive comparison lets you test these calls against any metro area.