Twelfth in a series on jobs whose pay system is stranger than the salary. Earlier entries covered coal and metal miners, longshoremen, and harbor pilots, each a case where the interesting money sits off to the side of the wage. Oilfield hands belong next to the miners, because both are extraction jobs, but the oilfield adds a device the coal mine does not have: the entire wage structure is bolted to the price of a commodity, and when the commodity falls, the jobs vanish by the tens of thousands. Every figure below is cited to a federal wage series, a Federal Reserve data series, an industry report, or a job posting, and where a number could not be verified at a primary source I say so rather than estimating.
The median wage for an oilfield hand tells you almost nothing about what the hand actually earns.
Not because the number is wrong. The federal survey says an entry-level roustabout has a median of $46,960 a year, and that is a real, defensible figure. The problem is that the survey measures a wage, and an oilfield worker is not really paid a wage. The pay is engineered around two things a salary chart cannot see: the oil price and the rig schedule. A 40-hour week is not the unit of work here. The unit is the hitch.
Four structural devices stack on top of that plain median, and none of them appear in it. First, the hitch itself, a fixed block of days lived on the rig pulling one 12-hour tour every single day, which turns the "wage" into base pay plus a wall of overtime crammed into roughly six months of the year. Second, the day rate, a flat dollars-per-day the senior field roles are paid, often as 1099 contractors, so their income is rig-days times a number rather than a salary. Third, per diem and, in booms, cash signing bonuses, a large and tax-advantaged slice of take-home that never touches the base rate. And fourth, the one that governs all the others, boom-bust volatility. The same job can pay six figures with overtime and per diem in a good year and simply not exist in a bad one.
Take them in order, but start with the wage the survey does report, because the ladder inside it is the first tell.
What the survey says the job pays
The federal wage program groups the field into four production occupations, and they form a clean ladder from the ground up. These are the May 2025 medians as republished by O*NET OnLine, the Labor Department's own occupational database, which reproduces the Bureau of Labor Statistics wage series. I am labeling them secondary for a specific reason given at the end: bls.gov itself blocked every automated route I tried, so I read the BLS numbers through the DOL-sponsored mirror rather than the source page.
| Oilfield production occupation (SOC) | Median annual | Median hourly | Employed |
|---|---|---|---|
| Roustabouts, 47-5071 | $46,960 | $22.58 | ~46,000 |
| Service unit operators, 47-5013 | $58,160 | $27.96 | ~45,200 |
| Derrick operators, 47-5011 | $58,620 | $28.18 | ~11,300 |
| Rotary drill operators, 47-5012 | $67,890 | $32.64 | ~13,300 |
The roustabout is the general laborer, the person who does the physical work on and around the rig. The service unit operator runs the well-servicing and workover equipment that keeps existing wells producing. The derrick operator works up in the mast handling pipe. The rotary drill operator, usually just called the driller, runs the rig floor and is the senior hand of the four. The ladder tops out around $68,000, which is a solid blue-collar wage and nothing exotic.
The Bureau's own umbrella figure for the four occupations combined lands right in the middle of that ladder. The Occupational Outlook Handbook reports a median of $52,610 for "oil and gas workers" as a group in May 2024, and puts the group at about 115,900 jobs in 2024. One honest note on vintages: the four occupation medians in the table are the May 2025 survey, while that $52,610 umbrella figure is May 2024, so they are one year apart and should not be added together or averaged. I am keeping them clearly separated for that reason.
So on the survey alone, this looks like an ordinary trade. It is not. Everything that makes it strange is bolted on top of these numbers, and the first bolt is the schedule.
Device one: you live on the rig, and the week is 84 hours
An oilfield worker does not commute to a job. The worker deploys to it for a fixed block of days called a hitch, lives there for the whole block, then goes home for another block.
The rig runs two 12-hour tours a day, a day tour and a night tour, and a hand works one of them every day of the hitch. Not five days. Every day. So a seven-day stretch is an 84-hour week, and the whole hitch is worked at that pace before anyone goes home. Offshore, the common rotations are 14-and-14, 21-and-21, or 28-and-28, meaning that many days on followed by an equal block off, and North Sea norms run to 14-on with 21-off or a straight 28-and-28. Onshore drilling hitches more often run 7-and-7 or 14-and-14.
Now do the arithmetic that the recruiters do. On an even-time rotation, equal days on and off, a worker is physically on site only about 182 days a year. Half the calendar is time off. But the annual pay is compressed into the working half, at roughly 84 hours a week, which means it is built almost entirely out of overtime. The base hourly rate on the survey is real, but a hand earning it works so far past 40 hours, for so many consecutive days, that the base is only the seed. The overtime is the tree.
This inverts the trick from earlier in this series. The longshore contract pays for more hours than are worked, through guarantees. The oilfield does the opposite and pays for every one of a great many hours actually worked, by structuring the job so that a normal year of labor gets done in about six months of very long weeks. Either way, the annual number detaches from the hourly wage, and the salary chart never sees the gap.
One caution on the rotation figures. The specific hitch patterns above come from industry recruiter guides, not a government survey, so treat them as descriptive of common practice rather than a legal standard, and the "half the year off" figure is my own arithmetic off an even-time rotation, not a published statistic.
Device two: the senior hands are paid by the day, not the year
Climb above the crew and the pay stops even pretending to be a salary. It becomes a day rate.
The tool pusher, who runs the rig for the drilling contractor, and the company man, also called the drilling consultant, who represents the operator that owns the well, are frequently paid a flat dollar amount per day worked, and very often as 1099 independent contractors rather than employees. Their income is simply rig-days times the rate. No rig-days, no income. It is the purest possible version of pay-for-availability, and it is why the senior field roles are the ones most exposed to a downturn even though they earn the most in a boom.
I want to be careful with the numbers here, because this is the softest part of the record. Industry aggregator pages put onshore drilling-consultant and company-man day rates somewhere around $900 to $2,200 a day, tool pushers at roughly $80,000 to $200,000 a year, and rig managers and company men at $150,000 to $300,000 or more. Those are composite estimates from a salary-aggregator site, they read as generated rather than surveyed, and I could not confirm any of them against a staffing firm's published rate sheet or an industry compensation report. So I am reporting the shape of the structure, which is well established, day rate and often 1099, and flagging the specific dollar bands as unverified. If you need a real number, get it from a staffing firm's rate sheet, not from me and not from an aggregator.
The structural point survives the soft numbers. At the top of the field, compensation is decoupled from the calendar entirely. You are worth your day rate times the days the rig turns, and the rig turns only when the price supports it.
Device three: per diem and the boom bonus
Below the day rate, at crew level, two more things ride alongside the base wage, and both are built to move money without moving the posted hourly rate.
The first is per diem, a daily allowance for being away from home that is treated more favorably than wages and lands on top of the hourly pay. A single illustrative 2022 drilling-rig-hand posting advertised $25 to $36 an hour plus $50 a day per diem and housing. That is one job board listing, not a statistical source, and per diem amounts swing widely by basin, company, and year, so I offer it only as a shape, not a figure to bank on. But the mechanism is standard across the field: part of take-home is per diem, and the per diem is not in the wage line.
The second appears only in booms. When crude spikes and rigs come back faster than crews can be hired, companies compete for scarce hands with cash signing bonuses. The labor-shortage narrative of the 2022 Permian boom is well known anecdotally. I could not find a primary, quantified source for bonus sizes, no Dallas Fed survey figure or company disclosure I could stand behind, so I am flagging the bonuses as real in kind and unverified in amount. I will not print a bonus number I cannot source.
Put the three devices together and the picture is a paycheck assembled from parts that a salary chart is structurally unable to add up: a base wage, a wall of overtime from the hitch, a day rate at the top, a per diem on the side, and, in a good year, a bonus on signing. And then the fourth device takes all of it away.
Device four: the whole structure tracks the oil price, and the evidence is in the headcount
This is the device that makes the oilfield stranger than the coal mine, and it is the one part of this article resting on primary, verified data rather than industry description.
When the oil price collapses, these jobs do not get a pay cut. They disappear. And because the federal employment series count the people doing this work, you can watch the workforce get shed in near real time. The cleanest proxy is Support Activities for Mining, the industry bucket where most oilfield-services, drilling-contractor, and well-servicing crews are counted. Here is what the Federal Reserve's copy of the Bureau of Labor Statistics headcount series shows across the last two busts.
| Support Activities for Mining, employment | Level |
|---|---|
| Peak, September 2014 | 444,900 |
| Trough, August 2016 | 254,300 |
| Recovery, January 2019 | 360,200 |
| Trough, February 2021 | 204,800 |
| Most recent, June 2026 | 263,700 |
The 2014 to 2016 collapse erased about 190,600 jobs, roughly 43 percent of the workforce, tracking the oil-price crash almost directly. The sector clawed back to 360,200 by early 2019, then the 2020 pandemic and price war knocked it down to 204,800 by February 2021, a further drop of about 43 percent from the 2019 peak and roughly 54 percent below the 2014 high. And the recovery is still not complete. As of June 2026, employment stood at 263,700, which is still about 41 percent below where it was in September 2014. More than a decade later, the field has not gotten its jobs back.
A narrower series tells the same story. Oil and Gas Extraction employment peaked at 200,800 in October 2014 and fell to a trough of 110,900 in November 2021, a decline of roughly 45 percent.
If you want the leading indicator behind those employment swings, it is the rig count, because rigs are what crews staff. The U.S. rotary rig count, from Baker Hughes' weekly tally, bottomed at 404 rigs in late May 2016, down from a peak above 1,900 in the autumn of 2014. In the 2020 bust it fell to 244 in mid-August 2020, widely reported as the lowest since Baker Hughes began issuing weekly U.S. numbers in 1944. Rigs get stacked, and when a rig stacks, its crew is laid off, its tool pusher's day rate goes to zero, and its company man goes home. The rig count is the payroll.
Two honesty notes on this section. The employment levels are primary and verified, pulled from the exact BLS establishment-survey series. The rig-count figures are secondary, reported through a trade outlet quoting Baker Hughes rather than read from Baker Hughes' own dataset, and the "above 1,900" autumn-2014 peak is approximate: I saw it widely cited around 1,920 to 1,931 rigs but did not confirm a specific number, so I am not printing one.
The white-collar contrast on the same well
There is one more figure worth setting against the field crews, because it shows how the same barrel of oil pays two completely different pay systems.
The petroleum engineer designs the well the crews drill. The median for that occupation is $144,910 a year, or $69.67 an hour, in the May 2025 survey as republished by O*NET, against about 19,600 employed. The Occupational Outlook Handbook's May 2024 figure is $141,280, slightly lower and one vintage older; I am citing both and treating $144,910 as the newer number. Either way, the engineer's median is more than double the driller's $67,890 and roughly three times the roustabout's $46,960.
The engineer is salaried, office-based, and largely insulated from the hitch. When the driller is pulling an 84-hour week offshore or getting laid off in a bust, the engineer is drawing a stable annual salary. Same commodity, same well, two pay systems that could not be more different in their exposure to the oil price. The lesson from the harbor pilots and longshoremen holds here too: whether a job's pay is stable or volatile has almost nothing to do with how hard or skilled the work is, and almost everything to do with the structure wrapped around it.
What a salaried reader should take from this
When pay is compressed into part of the year, the annual number and the hourly wage stop matching. A roustabout's $22.58 an hour and a driller's $32.64 look modest until you remember the hitch. The worker earns a full year's living in roughly six months of 84-hour weeks, so annual take-home is mostly overtime, per diem, and, at the top, a day rate. Any time you compare a job to yours, ask whether it is paid across the calendar the way a salary is, or crammed into a fraction of it. The two produce very different lives from the same headline wage.
A day rate is not a salary, and 1099 is not W-2. The senior field roles are paid by the day, often as contractors, which means no rig-days is no income, no employer benefits, and no floor under a bad year. That structure buys enormous upside in a boom and offers zero protection in a bust. Whenever an offer is quoted per day or per project rather than per year, the real question is how many of those units you can count on when the market turns against you.
The pay system and the number of jobs move together, and both track the commodity. This is the device with no equivalent in most work. When crude fell, oilfield employment did not shrink by a few percent through attrition. It fell about 43 percent in two years, twice, and a decade after the 2014 peak it is still down roughly 41 percent. The coal miner's pay has a federal disease tax bolted to it; the oilfield hand's pay has the oil price bolted to it, and the oil price can end the job entirely. If your income depends on a single commodity or a single customer's capital budget, the relevant risk is not a pay cut. It is the whole job disappearing on a price move you do not control.
The figure you can verify is often not the figure that matters. The one number in this article resting on primary, audited-grade data is the headcount, and it is the number that decides everything, because it is the count of who still has the job. The day rates, the per diem, the bonuses, the pieces that make the boom feel rich, are exactly the pieces I could not verify at a primary source. That is worth noticing in your own pay too. The parts that are easy to look up are rarely the parts that determine what a year actually pays.
Related reading
- How coal and metal miners are paid: the other extraction job, where a federal tax on the disease the work causes sits beside the wage.
- How harbor pilots are paid: a job whose income is set by a state formula rather than a market, and whose federal wage category cannot even see the people doing it.
- How longshoremen are paid: a modest base wage multiplied into a six-figure average by hours rules a salary chart never shows.
- How railroad workers are paid: paid by the mile, on a schedule that has little to do with a 40-hour week.
- COLA versus the merit raise: why what your pay is tied to matters more than where it starts.
Fact-check notes and sources
Wage figures come from the Bureau of Labor Statistics wage program as republished by the Labor Department's O*NET database and the Occupational Outlook Handbook. Employment figures come from the BLS establishment survey as hosted by the Federal Reserve. Rig counts, day rates, per diem, and bonuses come from industry and job-board sources of varying reliability, and each is labeled accordingly. Where a figure could not be verified at a primary source, it is flagged in the text and here.
- The four field occupation medians are the BLS Occupational Employment and Wage Statistics May 2025 release as republished by ONET OnLine: roustabouts, SOC 47-5071, at $46,960 a year and about 46,000 employed, from O*NET 47-5071.00; service unit operators, SOC 47-5013, at $58,160 and about 45,200, from O*NET 47-5013.00; derrick operators, SOC 47-5011, at $58,620 and about 11,300, from O*NET 47-5011.00; and rotary drill operators, SOC 47-5012, at $67,890 and about 13,300, from O*NET 47-5012.00. These are secondary: bls.gov itself returned an Akamai "Access Denied" or 403 to every automated fetch route I tried, including direct requests and two proxy paths, so the BLS numbers were read through the DOL-sponsored ONET mirror rather than the origin pages. To cite BLS directly, retry these via a residential or browser session: bls.gov/oes/current/oes475011.htm, oes475012.htm, oes475013.htm, and oes475071.htm.
- Petroleum engineers, SOC 17-2171, at a median of $144,910 a year and about 19,600 employed in the BLS OEWS May 2025 release, are from O*NET 17-2171.00. The May 2024 Occupational Outlook Handbook figure of $141,280 for the same occupation is from the BLS OOH, Petroleum Engineers (page Akamai-blocked to automated fetch; figure verified via search result). The two figures are one survey vintage apart and are labeled as such.
- The umbrella "oil and gas workers" median of $52,610 in May 2024, and the group total of about 115,900 jobs in 2024, are from the BLS Occupational Outlook Handbook, Oil and Gas Workers. This page is Akamai-blocked to automated fetch; the figure was obtained via the BLS search result and is labeled secondary. Note the vintage difference: this umbrella number is May 2024, while the four occupation medians above are May 2025.
- Employment in Support Activities for Mining (NAICS 213), where most oilfield-services, drilling-contractor, and well-servicing crews are counted, is the BLS establishment-survey series CES1021300001 as hosted by the Federal Reserve: the peak of 444,900 in September 2014, the trough of 254,300 in August 2016, the recovery to 360,200 in January 2019, the trough of 204,800 in February 2021, and the most recent 263,700 in June 2026, from FRED CES1021300001. These are primary and verified. The percentage declines are arithmetic from those levels.
- Employment in Oil and Gas Extraction (NAICS 211), peaking at 200,800 in October 2014 and reaching a trough of 110,900 in November 2021, is from FRED CES1021100001. Primary and verified.
- The U.S. rotary rig count lows, 404 rigs in late May 2016 and 244 in mid-August 2020 (the latter widely reported as the lowest since Baker Hughes began weekly U.S. numbers in 1944), are the well-known figures from the Baker Hughes North America Rotary Rig Count. Secondary, and a supporting detail rather than the load-bearing evidence: these specific weekly figures are widely reported but I did not read them from the Baker Hughes dataset directly, so the boom-bust case rests on the primary, verified employment series above, with the rig count as corroboration to confirm at the dataset. The autumn-2014 peak is approximate: it is widely cited around 1,920 to 1,931 rigs, but I did not confirm a specific number, so only "above 1,900" is asserted.
- The hitch and rotation structure, offshore rotations of 14-and-14, 21-and-21, or 28-and-28 with North Sea norms of 14-on and 21-off or 28-and-28, two 12-hour tours a day, and onshore hitches of 7-and-7 or 14-and-14, is from industry recruiter guides including Orion Group and iPS Powerful People. Secondary, describing common practice, not a legal standard. The "about 182 days on site" and the 84-hour week are my own arithmetic from an even-time rotation and a daily 12-hour tour, labeled as such, not published statistics.
- Day rates for senior field roles, onshore drilling-consultant and company-man rates around $900 to $2,200 a day, tool pushers at roughly $80,000 to $200,000 a year, and rig managers and company men at $150,000 to $300,000 or more, are from Rigzone salary-insight pages. Unverified: these are aggregator composite estimates that read as generated rather than surveyed, and I could not confirm them against a staffing firm's rate sheet or an industry compensation report. The day-rate and 1099 structure is well established; the specific dollar bands are not asserted as fact.
- Per diem and the illustrative hourly range, a 2022 drilling-rig-hand posting at $25 to $36 an hour plus $50 a day per diem and housing, is from a job-board posting. Unverified and illustrative only: it is a single listing, not a statistical source, and per diem and hourly rates vary widely by basin, company, and year. Boom signing bonuses are real in kind but I found no primary quantified source for their size, so no bonus figure is asserted.
- A primary BLS synthesis exists but could not be fetched. The BLS Monthly Labor Review article "Occupational employment and wages in oil and gas industries, May 2024" contains per-occupation medians and industry breakdowns, but the page returns an Akamai "Access Denied" to automated fetch and must be retrieved manually or via a residential proxy: BLS Monthly Labor Review. It is flagged here for retry rather than cited as read.
- The offshore-versus-onshore wage premium is qualitatively real, offshore day rates and totals run higher and remote-location premiums exist, but I could not obtain a primary BLS industry figure for the exact premium because the OEWS industry pages for NAICS 211 and 213 are Akamai-blocked; no premium figure is asserted.
This post is informational and journalistic, not career, legal, or financial advice. It describes government wage and employment data, a Federal Reserve data series, industry reports, and job-board postings. Wage figures are the May 2024 and May 2025 survey vintages as labeled, employment runs through June 2026, and market conditions in this industry change fast, so verify current status before relying on any of them. Mentions of specific companies, agencies, and industry sources are nominative fair use, and no affiliation is implied.
