Eighteenth in a series on jobs whose pay system is stranger than the salary. Earlier entries covered harbor pilots, whose income is a state board voting on how many people may hold the job, and longshoremen, whose wage is protected by a closed register. The taxi medallion owner belongs at the end of that line, because this is the scarcity mechanism stripped down to its purest form. There is no ship to guide and no cargo to move. There is only a license, capped by the government, that you lease to somebody else. It is also the one entry in this series where the toll booth collapsed, so it doubles as a warning. Every figure below is cited to a city agency report, a mayoral press release, or a Pulitzer citation, and where a number could not be verified at a primary source I say so rather than estimating.
Almost every job in this series pays well because somebody enforces scarcity. A state board caps the harbor pilots. A register caps the longshoremen. A flight-hour rule caps the airline pilots. But in each of those cases the scarce thing is a person doing work, and the pay is at least attached to labor performed.
The taxi medallion is what happens when you remove the labor entirely and keep only the scarcity.
A medallion is a metal plate. It is a city-issued license, capped by law, that permits one car to operate as a New York yellow cab. An owner who never drives a day in their life can buy the plate, hang it on a car, lease that car to a driver who does the actual work, and collect rent. The driver earns the fares. The owner earns the scarcity. This is the toll booth with the pretense removed, and for most of a century it was one of the best passive assets an ordinary immigrant could buy.
Then the city that manufactured the scarcity erased it, and the asset that thousands of people had mortgaged their lives to own became a debt that outlived its own value.
The number the city froze
The whole thing rests on one number that the government agreed not to increase.
The count of New York yellow-cab medallions is capped by law. It sat at 11,787 for decades, unchanged, until the City ran its first new auctions in 1996 and 1997. Today the authorized figure is 13,587, and no new medallions have been auctioned since 2014. That is the entire supply. It does not grow with the population, the tourism, or the demand for a ride. It is a fixed integer, and a fixed integer is exactly what makes a license valuable, because everyone who wants in has to buy their way in from someone already holding one.
Watch what that scarcity did to the price over seventy years:
| Year | Approximate medallion price |
|---|---|
| Post-WWII | $2,500 |
| 1961 | $21,000 to $23,000 |
| 1985 | $100,000 |
| 2011 | first exceeded $1 million |
| 2013 | peak of about $1.3 million |
A license that cost $2,500 after the war reached roughly $1.3 million at its 2013 peak. Nothing about the physical job changed. A cab in 2013 did the same thing a cab did in 1948. What changed was that the population and the demand kept climbing while the number of plates stayed frozen, so the plate itself became the asset, and it behaved less like a work permit than like a bond that happened to come with a car attached.
How a non-driving owner actually gets paid
The city's own task force sorts medallion owners into three groups, and the sorting is the clearest window into how the money moves.
There are active fleet managers, who own medallions and lease them directly to drivers. There are passive investors, who own medallions and hand them to a management company to lease out on their behalf. And there are owner-drivers, who operate their own cab. The first two groups never have to touch a steering wheel. They collect lease income, which is rent on the license, and that rent is the pay.
The amount an owner can charge a driver is not a free-market number either. The Taxi and Limousine Commission caps it, along with how long a driver may lease the vehicle or the medallion, under what the rules call the lease-cap rules. So even the rent on the scarce asset is itself regulated. I want to be honest about a limit here: the current dollar figure of a daily or weekly lease cap comes from a TLC rule I did not retrieve at its primary source, so I am describing the structure, that lease amounts are capped by the agency, without asserting a present-day lease rate. Press accounts often cite figures on the order of $100 or more per shift, and I could not verify those at a primary source, so I do not restate them as fact.
The shape is what matters. A salaried worker is paid for hours. A medallion owner is paid for holding a piece of paper that the city refuses to print more of. The driver in the seat generates the fares and keeps what is left after the lease. The owner keeps the scarcity.
What the scarcity was worth at the top
The clearest proof that people were buying the license and not the labor is what they paid for it in the final auctions, right before the collapse.
Between November 2013 and March 2014 the TLC auctioned 400 medallions across three auctions, raising about $359 million for the City. The highest price paid for an independent medallion was $965,000, and the average across that class was $863,742. In the March 2014 auction, the last TLC auction to date, a single lot of two wheelchair-accessible corporate medallions drew a high bid of $2,420,500, which is $1,210,250 for each plate. The average in that corporate class was $2,328,757, or $1,164,378 each.
Sit with those figures. A person bid more than $1.2 million for one license to operate one taxi. Nobody pays that for a job. They pay it for an asset they expect to appreciate, or at least hold its value, because the government has promised, implicitly, to keep the supply frozen. The city was happy to sell that promise. It took in more than $855 million from medallion auctions and transfer taxes between 2002 and 2014, with about $359 million of that from the final auctions alone.
When the toll booth lost its toll
A toll booth is only worth something if drivers have to pass through it. The medallion's entire value depended on the yellow cab being the only legal way to hail a ride in New York. Then the city licensed an alternative and put no cap on it at all.
Until 2018 there was no limit on the number of app-based, for-hire vehicles allowed to operate in the city. The contrast with the frozen medallion count is the whole story. On one side of the street, 13,587 licenses, fixed by law. On the other side, an open door. By March 2018 the TLC had about 130,000 active licensed for-hire vehicles and was issuing licenses for roughly 2,000 new vehicles every month. Active TLC licenses of all kinds soared to 343,786 by 2018.
The riders followed the supply. Medallion taxis averaged about 485,000 trips a day in 2014. By early 2018 that had fallen to roughly 296,000 a day, while the app-based companies averaged about 600,000 trips a day. In March 2019 alone, high-volume for-hire services accounted for at least 24 million trips across the city. The yellow cab had not gotten worse. It had simply lost its monopoly on the curb, and the monopoly was the product.
So the price of the license did what the price of any asset does when the scarcity behind it evaporates:
| Date | Average medallion sale price |
|---|---|
| 2013 peak | about $1.3 million |
| January 2018 | $188,796 |
| November 2019 | $164,518 (median $200,000) |
By 2019, 66 percent of the medallion transfers that took place were not sales at all. They were foreclosures. The market for the asset had become a market for repossessing it.
The debt that outlived the asset
Here is where the pay system turns cruel, and where it stops being a story about clever passive income and becomes a story about leverage.
Most of the people holding these licenses were not Wall Street investors who could absorb a loss. Approximately 95 percent of all medallion taxi drivers are immigrants, and most licensed TLC drivers do not speak English as their primary language. Many of them were owner-drivers who had borrowed nearly the entire purchase price of the medallion, because when a license costs $1 million and reliably appreciates, borrowing against it looks like the safest bet in the world. Until it is not.
When the price collapsed, the debt did not. A loan written against a $1 million medallion is still a loan against $1 million even after the medallion is worth $180,000. In a set of more than 300 court cases from 2017 to 2019, covering about 400 loans and roughly $395 million of debt on about 560 medallions, the average outstanding indebtedness was about $701,000 per medallion. In a TLC survey of about 450 owner-drivers, the median outstanding debt was about $499,000, with a median monthly payment of about $2,600. More than 950 medallion owners filed for bankruptcy, and the task force described thousands more as barely hanging on.
The loans themselves had been built to hide the risk. Most featured large balloon payments. Some required interest-only payments, meaning no principal was ever repaid, and The New York Times reported that some of those interest-only borrowers did not realize their principal was not being paid down at all. The federal credit-union regulator later found that lenders frequently based their decisions on inflated market values of the medallions rather than on accepted best practices, producing loans that the underlying cash flow could never support.
Two people from the Times reporting make the arithmetic concrete. Mohammed Hoque, a Queens owner-driver whose annual income was around $30,000, bought a medallion in 2014 under a loan that required him to repay $1.7 million. He drove six days a week, 12 hours a day, still fell behind, and had his medallion repossessed at least six times. Yvon Augustin made interest-only payments of $2,275 a month, watched none of it reduce what he owed, and eventually declared bankruptcy and lost his medallion.
The lenders went down with the borrowers. Three credit unions that had concentrated heavily in medallion loans, Melrose Credit Union, LOMTO Federal Credit Union, and Bay Ridge Federal Credit Union, became insolvent and were taken over by the National Credit Union Administration. By the time they failed they collectively held about $1 billion in medallion loans, and their insolvencies caused an estimated $765.5 million loss to the federal share-insurance fund. And the human cost ran past dollars. At least eight New York taxi and for-hire drivers died by suicide amid the debt crisis, some of them medallion owners with loans they could not survive. Reporting varies on the exact tally, with some later accounts citing nine, and I could not reconcile the count against a single primary source, so I report the range and flag it.
The reckoning, the bailout, and the Pulitzer
The scale of it eventually forced two things that rarely happen: an investigation that changed the record, and a public bailout of a private asset.
The investigation was Brian M. Rosenthal's series for The New York Times, which won the 2020 Pulitzer Prize for Investigative Reporting. Over ten months the Times interviewed 450 people, built a database of every medallion sale since 1995, and reviewed thousands of loans and documents, exposing how a small group of industry leaders had inflated medallion prices and funneled drivers into loans they could not repay. That reporting is the spine of the public record on this collapse, and much of what the city's own task force later documented traces back to it.
The state moved too. New York Attorney General Letitia James alleged that the TLC had run a scheme that inflated medallion prices, and she sought roughly $810 million on behalf of buyers. I want to be careful here: that claim and its resolution come from news coverage, not a primary court or attorney-general document I retrieved, so I report the allegation and not an outcome.
The bailout came in 2022. Under the finalized Medallion Relief Program Plus, announced on August 30, 2022, the lender Marblegate Asset Management agreed to restructure the outstanding loans of more than 3,000 medallion owners. The mechanics: a $30,000 city grant is applied as a down payment, the loan is reduced to a $200,000 base, and the restructured principal is capped at a maximum of $170,000. The city funds a loan guarantee and loss reserve behind it. The restructured loans carry interest of 7.3 percent or less, are fully amortized over no more than 25 years, and cap the monthly payment at $1,234. Owners could begin closing on the new terms starting September 19, 2022. A public program, funded by taxpayers, stepped in to repair the damage from a private license whose value the same government had first inflated and then destroyed. How many drivers have actually closed on the restructured loans, and the total debt ultimately written down, are post-2022 outcome figures I could not confirm at a primary TLC or comptroller source, so I do not assert them.
The medallion is not worthless today. Prices have partially stabilized well below the peak, reported roughly in the range of $90,000 to $200,000, with commonly cited 2024 averages around $125,000 to $170,000, and transfer volume rose from 777 in 2023 to about 1,003 in 2024, with the relief program credited for clearing the mass-foreclosure overhang. Those current-price figures come from a City Council hearing transcript and secondary market trackers rather than a downloaded primary TLC dataset, so treat them as approximate. And the license still carries a public accessibility obligation on its face, with a large share of the fleet required to be wheelchair-accessible. The figure widely cited is half of the 13,587 authorized medallions, or 6,794, by the end of 2028, but I could not confirm that exact 50 percent target and deadline at the TLC yellow-cab page, which states only the 13,587 total, so treat the specific accessibility figure as reported rather than verified.
What a salaried reader should take from this
The purest passive income is a toll on scarcity, and the scarcity may not be yours to keep. A medallion owner who never drove was paid for holding a license the city refused to duplicate. That is the same engine behind the harbor pilots and the longshoremen, scarcity that somebody else enforces. But the medallion exposes the catch the others hide: when a government manufactures the scarcity, the government can un-manufacture it. The pilots' cap is defended by a public board that votes down the shipping lobby. The medallion's cap was undone by the same city that sold it, the day it licensed an uncapped competitor across the street.
A capped license behaves like a bond until the cap moves, and then it behaves like nothing. For seventy years the medallion only went up, so borrowing the full price against it looked riskless. The risk was never in the cash flow. It was in a single policy decision that could, and did, erase the moat overnight. Any asset whose entire value is a rule can be revalued to near zero by a change in that rule. If your net worth rests on a license, a permit, or a franchise that a public body could reissue or expand, you own a policy, not a property.
Leverage priced off a toll survives the toll. The people who bought medallions with cash lost value. The people who borrowed against them lost their lives' savings, because a $700,000 loan is still $700,000 after the collateral falls to $180,000. Debt secured by a scarce license does not fall when the scarcity does. That asymmetry, the asset can collapse but the loan cannot, is the mechanism that turned a good passive investment into more than 950 bankruptcies. Borrowing to buy a scarcity is a bet that the scarcity is permanent, and no scarcity created by rule ever is.
When one authority both prints the license and profits from selling it, read the incentive. The city capped the supply, auctioned the plates, collected more than $855 million in auction and transfer revenue, and then licensed the competition that destroyed the value. That is not a claim of bad faith, it is a structural observation, the same one that runs through the COLA piece: the party that sets the scarcity is rarely a neutral party, and its interests and yours can diverge sharply at exactly the moment you can least afford it.
Related reading
- How harbor pilots are paid: the same scarcity engine defended by a public board that votes against the industry lobby, which is why that toll booth still stands.
- How longshoremen are paid: a closed register that keeps the wage from being competed down, the union version of a capped license.
- How alcohol distributors are paid: another state-manufactured franchise where the license, not the labor, is the asset.
- The raise written into the contract: the wider set of workers whose gains come from a rule someone else has to honor.
- COLA versus the merit raise: why a number set by a rule pulls away from one set by a market, in both directions.
Fact-check notes and sources
Nearly every hard figure in this article comes from a single primary document, the New York City Council's own task force report, corroborated where possible by the mayor's office and the Pulitzer board. Where a figure is secondary or could not be verified, it is flagged in the text and here.
- The medallion cap and its history (13,587 authorized medallions; the historical count of 11,787 held for decades until the 1996 and 1997 auctions; no new medallions auctioned since 2014), the three owner types and the TLC lease-cap rules, the long-run price history (about $2,500 post-WWII, $21,000 to $23,000 by 1961, about $100,000 by 1985, first over $1 million in 2011), the peak of about $1.3 million in 2013, the 2013 to 2014 auction results (400 medallions across three auctions raising about $359 million; independent high of $965,000 and average $863,742; the March 2014 corporate lot at a $2,420,500 high bid, or $1,210,250 each, average $2,328,757, or $1,164,378 each), the collapse figures (January 2018 average $188,796; November 2019 average $164,518 and median $200,000), the 66 percent foreclosure share of 2019 transfers, the uncapped for-hire competition (no cap before 2018; about 130,000 active licensed for-hire vehicles by March 2018; roughly 2,000 new licenses a month; 343,786 active TLC licenses by 2018), the ridership shift (about 485,000 medallion trips a day in 2014 falling to roughly 296,000 by early 2018; app-based averaging about 600,000 a day; at least 24 million high-volume for-hire trips in March 2019), the 95 percent immigrant share of drivers, the debt figures (more than 950 bankruptcies; the 300-plus court cases covering about 400 loans and roughly $395 million on about 560 medallions at an average of about $701,000 each; the TLC survey median debt of about $499,000 and median monthly payment of about $2,600), the predatory loan structures (balloon and interest-only loans, borrowers unaware principal was not being repaid, lending on inflated values per the NCUA review), the lender collapse (Melrose, LOMTO, and Bay Ridge credit unions taken over by the NCUA holding about $1 billion in medallion loans, with an estimated $765.5 million loss to the federal share-insurance fund), the individual cases of Mohammed Hoque (about $30,000 income, a 2014 loan requiring repayment of $1.7 million, driving six days a week and 12 hours a day, repossessed at least six times) and Yvon Augustin (interest-only payments of $2,275 a month before bankruptcy), and the auction and transfer revenue (more than $855 million between 2002 and 2014) are all from the New York City Council Taxi Medallion Task Force Report, January 2020: Taxi Medallion Task Force Report (Final).
- The Medallion Relief Program Plus terms (announced finalized August 30, 2022; more than 3,000 owners; Marblegate restructuring loans to a maximum principal of $170,000 after a $30,000 city-grant down payment reducing the loan to a $200,000 base; a city-funded loan guarantee and loss reserve; interest of 7.3 percent or less; amortization over no more than 25 years; monthly payments capped at $1,234; closings beginning September 19, 2022) are from the NYC Mayor's Office press release, Mayor Adams, TLC, Marblegate, NYTWA Announce Historic Taxi Medallion Debt Relief Program Deal, August 30, 2022. The number of drivers who have actually closed and the total debt written down are post-2022 outcome figures not confirmed at a primary source and are not asserted here.
- The Pulitzer-winning investigation (Brian M. Rosenthal, The New York Times, 2020 Pulitzer Prize for Investigative Reporting; the ten-month reporting, 450 people interviewed, a database of every medallion sale since 1995) is from The Pulitzer Prizes citation, which is a secondary but authoritative source; the underlying Times reporting on driver suicides is at The New York Times. The exact tally of driver suicides could not be confirmed at a single primary source; reporting cites at least eight, with some later accounts saying nine, and the range is flagged in the text rather than resolved.
- The Attorney General's allegation that the TLC inflated medallion prices and the roughly $810 million sought on behalf of buyers is from news coverage, Gothamist, and is secondary; the auction and transfer-revenue figures it references are corroborated by the task force report above. The final resolution of that action was not confirmed at a primary court or Office of the Attorney General source and is not asserted.
- Current (2024 to 2025) medallion prices (roughly $90,000 to $200,000, with commonly cited 2024 averages of about $125,000 to $170,000) and the rise in transfer volume (from 777 in 2023 to about 1,003 in 2024) are from New York City Council Committee on Transportation testimony, February 10, 2025, and secondary market trackers, and are treated as approximate rather than a downloaded primary TLC transfer-price dataset.
- The accessibility mandate (the widely cited target of half of the 13,587 authorized medallions, or 6,794, attached to wheelchair-accessible vehicles by the end of 2028) is referenced against NYC TLC Yellow Cab materials, nyc.gov/site/tlc. That page confirms only the 13,587 total, not the 50 percent target or the 2028 deadline, so the specific accessibility figure is flagged as reported rather than verified and would need confirmation against the underlying TLC accessibility rule (35 RCNY / WAV dispatch rules).
- Not verified and therefore not asserted as fact: the current dollar figure of the TLC lease cap an owner may charge a driver (the task force confirms the cap exists under the lease-cap rules, but the present-day schedule from 35 RCNY was not retrieved, and press figures on the order of $100 or more per shift were not confirmed at a primary source); the share of the 13,587 medallions currently encumbered by debt (the task force states this could not be determined because most loan transactions are not public record); and whether the "$1.3 million in 2013 peak" is a market-wide figure versus a characterization, since the hardest individual data points are the auction highs of 2014.
This post is informational and journalistic, not career, legal, or financial advice. It describes a published city council report, a mayoral press release, a Pulitzer Prize citation, and news coverage. Medallion prices, program terms, and rules change, and several figures are as of 2019 through 2025 as noted, so verify current status before relying on any of them. Mentions of specific agencies, companies, lenders, and named individuals are drawn from the public record and are nominative fair use, and no affiliation is implied.
