Editorial Article

The 46.70 Per Cent Was the Discount Rate: Chicago's Meter Lease Reproduces to Four Tenths of a Point, and That Is the Worse Finding

Updated November 20, 2026

In December 2008 the City of Chicago leased its 36,000 metered parking spaces for seventy-five years and received USD 1.157 billion. The city’s own inspector general, reporting in 2009, put the value of the same revenue stream at USD 2.154 billion. The distance between the two is USD 997 million — 46.29 per cent — and for seventeen years it has been the standard exhibit in every argument about public asset monetisation.

The usual reading is that Chicago was out-negotiated. This article proposes a different one, and it is not a defence of the deal. Discount the identical cash flow — seventy-five years, growing at 2.5 per cent a year — at the 5 per cent a tax-exempt municipal borrower pays and at the 8 per cent a taxable private buyer pays, and the private price comes out at 53.30 per cent of the municipal value. That is a shortfall of 46.70 per cent. The inspector general measured 46.29. The two numbers agree to 0.41 of a percentage point.

That agreement is worth more than an alibi. A negotiating failure is a lesson: hire better advisers, run a longer auction, publish the bid. A tax exemption is a structure, and structures do not respond to better behaviour. If the arithmetic reproduces the observed outcome to four tenths of a point, the honest conclusion is that a city selling a long-dated revenue stream is selling into a market that prices the asset roughly half again lower than the city is able to value it — and the only way to close that is to stop being the seller. The full modelling sits in the parent report on parking concession monetisation and revenue-stream financing; what follows is the part of it that a treasurer can test on one page.

What the inspector general actually measured

The comparison is easy to mis-state, so it is worth putting the units side by side. Chicago’s receipt divided across 36,000 meters is USD 32,138.89 per meter. The inspector general’s valuation divided the same way is USD 59,833.33. The per-meter difference is USD 27,694.44, and multiplied back out it is the USD 997 million headline.

Chicago meter lease, 2008 Per meter (USD) Total (USD)
Price received by the city 32,138.89 1,157,000,000
Inspector general’s valuation 59,833.33 2,154,000,000
Difference 27,694.44 997,000,000
Shortfall 46.29 per cent

Two things follow from that table. The first is that neither figure is a fact about the world; both are valuations, and they differ because they embed different assumptions. The city’s number is a price — what one buyer group, in one auction, in the worst quarter of the worst financial crisis in seventy years, was willing to pay. The inspector general’s number is a model. Preferring the model to the price is defensible, but it is a judgement about assumptions rather than a discovery of a discrepancy in arithmetic.

The second is that the auction was not obviously thin. The concession drew competitive interest and the city ran an adviser-led process. What the inspector general’s analysis tested was whether the consideration matched a defensible valuation of the stream — and on the assumptions it used, it did not. That finding stands. The question worth asking is what those assumptions left out.

A parking meter is a growing annuity, and it is priced as one

A seventy-five-year right to charge for parking is a fixed-income instrument with a concrete slab attached. Its value is the present value of a stream that grows, at best, with the tariff and with the retail price index behind it. For a stream growing at 2.5 per cent a year the valuation factor is the sum of seventy-five discounted terms, and it is extremely sensitive to the rate in the denominator.

At a 5 per cent discount rate the factor is 33.4362: each dollar of first-year net revenue is worth USD 33.44. At 8 per cent it is 17.8211. Nothing about the meters changed between those two numbers. The tariff did not move, the enforcement did not change, the drivers did not drive less. A 300-basis-point difference in who owns the cash flow removed 46.70 per cent of its value.

Run that backwards and you get the figure the buyer was actually underwriting. USD 32,138.89 per meter divided by the 17.8211 factor implies USD 1,803.42 of first-year net revenue per meter. That is the number that had to be true for the price to make sense to a taxable buyer at 8 per cent — and every subsequent argument about meter rates in Chicago is downstream of it. The tariff is not the variable that set the price. It is the variable that had to be moved afterwards to make the price work.

Which is why the collection mechanism matters as much as the rate. A tariff nobody enforces is a discount, not a price: the on-street enforcement and digital permitting layer is what turns a posted rate into the collected revenue a buyer is willing to capitalise, and it is the first thing a concessionaire invests in after signing.

The four tenths of a point

A correction, because it belongs in the record. Earlier material in this series described the two figures as agreeing “to within 1.7 points.” They do not. 46.70 against 46.29 is 0.41 of a percentage point. The narrower gap strengthens the argument rather than weakening it — a reproduction that lands within half a point of an independent valuation is a tighter result than one that lands within two — but the earlier figure was wrong and is withdrawn.

What the reproduction does not do is prove causation. The spread did not cause the inspector general’s finding; the finding was made on its own assumptions and stands on them. What the reproduction establishes is narrower and, for anyone advising a council, more useful: no explanation beyond the discount rate is required to account for the observed gap. A reader who wants to argue that Chicago was also badly advised has to argue that the bad advice happened to cost almost exactly what the tax exemption cost. That is a coincidence rather than an argument.

It also does not capture what the agreement contained. The published concession carries obligations and protections — capital commitments, compensation when the city takes actions that reduce meter revenue, rate-setting provisions — and each of those was priced by the buyer and would move both valuations. A single-factor reproduction is a diagnostic, not a valuation. Anyone using it as the latter is overreading it.

The term premium has collapsed

The most consequential thing about the reproduction is what happens when the term varies. The gap between what a tax-exempt owner and a taxable buyer will pay is not constant in the length of the lease. It widens.

Stream growing 2.5 per cent a year Factor at 5 per cent Factor at 8 per cent Buyer’s price as a share of municipal value Shortfall
50-year concession 28.0109 16.8494 60.15 per cent 39.85 per cent
75-year concession 33.4362 17.8211 53.30 per cent 46.70 per cent

At fifty years the spread takes 39.85 per cent. At seventy-five it takes 46.70. Twenty-five further years of a city’s tariff, and the buyer’s price falls from 60.15 per cent of municipal value to 53.30. The seller is being asked to hand over time at precisely the point where time has become least valuable to the buyer.

The report states the same result from the buyer’s side. Extending a concession from fifty years to seventy-five adds 19.37 per cent to its value at a 5 per cent discount rate — and 5.77 per cent at 8. Anyone still advising a council to maximise the receipt by lengthening the term is working from a model in which money was cheap. It is not, and the term premium went with it. The same compression shows up between two private owners: at fifty years the identical stream is worth 28.0109 times year-one net revenue to a buyer funding at 5 per cent and 14.6735 times to a levered buyer at 9 — 1.9089 times more to the cheaper owner, for holding the same asset.

What this is pricing today

The mechanism is not a historical curiosity. Capital committed against long-dated parking contracts — concessions, leases and management agreements of ten years or more — was USD 119,846.4 million in 2025 and is USD 244,115.1 million in 2033 at 9.30 per cent: an increase of 103.69 per cent. The bays underneath grow from 53.02 million to 72.99 million, 4.08 per cent, an increase of 37.67 per cent. The value grows nearly three times faster than the thing being valued.

That difference is capital per bay, which rises from USD 2,260 to USD 3,345 — 48.01 per cent. It is construction cost, not tariff. The market’s growth story is therefore a story about how expensive it has become to build a bay, told by people who are buying the right to charge for one. The two halves of that sentence are not the same business, and the discount rate is what connects them.

Sensitivity does the rest. One hundred basis points removes 14.16 per cent of the value and requires a 16.50 per cent tariff rise at constant volume to stand still. Two hundred basis points removes 25.25 and requires 33.77. No operator in this market has that pricing power, and the ones who do are frequently forbidden from using it — which is why the pipeline has moved to jurisdictions where thirty-year terms are already accommodated in concession law. The ten operators this series discloses as holding the long-dated contracts are Indigo Group, Q-Park, Interparking, SP+, APCOA Parking, National Car Parks, Empark, Wilson Parking, LAZ Parking and Impark — and their reported portfolios are the evidence that the term, not the tariff, is what is being bought.

Where the asset meets the curb, the same arithmetic applies to a different instrument. The curb and kerbside management market is a schedule rather than a car park, but it is sold on the same long-dated basis and priced by the same denominator. The transaction layer matters too: access and revenue control is sold by the transaction now, which turns a capital asset into a metered service and changes what can be financed against it, and mobile payment platforms take a fee on a session with no cost base of its own. Underneath all of it sits the same question: whether the thing being sold is a bay, a contract, or a right to charge — and the concession operations research in this series makes the case that only the second of those three is an asset.

What the interviews say, and what they could not answer

This article rests on the primary programme described in the parent report: thirty-one semi-structured interviews across seven markets with concession directors, chief financial officers, infrastructure fund partners, municipal treasurers and transaction advisers; nine conference-floor interviews at industry and infrastructure-finance events; one closed roundtable with five operators and one toll-authority finance director; and forty-seven written survey responses from asset owners and operators, conducted between 2024 and 2026. Two findings in this series are ours and are labelled as such.

The first is the one this article is built on: the monetisation price gap observed in practice is arithmetically the tax-exempt-to-taxable discount-rate spread. We did not arrive at that by modelling. We arrived at it because a treasurer at a European authority described a bid as “about half” what the authority’s own capital cost implied, and the arithmetic later turned out to be 53.30 per cent. That is a field result that survived contact with a spreadsheet, which is the order we would prefer every finding to arrive in.

The second: the digital layer converts a concession into a services contract, and therefore shortens the asset that can be financed against it. If the revenue relationship is a session on a phone rather than a barrier on a ramp, the thing being securitised is a platform obligation, not a physical right, and lenders price it accordingly. Several operators described renegotiating term length to keep the financing profile they wanted after a technology replacement — a rational response to an irrational instrument.

What we could not get is the number that would settle the argument. No awarding authority we spoke to would disclose the discount rate it used to benchmark a bid. Four said the benchmark was a “cost of capital” figure set centrally by treasury and not released to the transaction team. That is a governance finding in its own right: the single input that dominates the valuation is, in a meaningful share of transactions, invisible to the people running the auction.

Sources

City of Chicago, Office of the Inspector General, analysis of the parking meter concession (2009). Road Traffic Regulation Act 1984, section 55. Internal Revenue Code sections 103 and 141, and Internal Revenue Service Rev. Proc. 2017-13. Directive 2014/23/EU on the award of concession contracts. Directive (EU) 2024/1275. Ministry of Housing, Communities and Local Government local government finance returns for 2024-25 as tabulated by the RAC Foundation. Rating agency research updates on Indigo Group and Interparking. The Metropolis Technologies and SP+ transaction completion announcement. EDHEC Infrastructure Institute, valuation study, June 2025. Lucknow Municipal Corporation draft parking concession agreement, as published via the National Institute of Urban Affairs. NHS England ERIC returns. Department of Health (Northern Ireland) statements and explanatory and financial memorandum. Wikidata, for entity identifiers. No third-party market-research publication is cited anywhere in this article.

Provenance. Figures for the market, the discount-rate factors, the term premium and operator financials are those of the parent report, Global Parking Concession Monetisation & Revenue-Stream Financing Market Report, Size, Share and Forecast 2026–2033: The Lease Is a Bond, by Vipul Pandey. Two model inputs are parameter choices held flat and disclosed there: the 55 per cent long-dated share of managed bays, and 5.02 per cent construction inflation to 2033.

Frequently Asked Questions

Did Chicago simply get a bad price for its parking meters?

The price was USD 1.157 billion against an inspector general valuation of USD 2.154 billion — a 46.29 per cent shortfall. What the discount-rate analysis adds is that the same 75-year stream, growing at 2.5 per cent, is worth 33.4362 times its first-year net revenue at the 5 per cent a tax-exempt city borrower pays and 17.8211 times at the 8 per cent a taxable buyer pays. That is a 46.70 per cent shortfall from the buyer's side alone, within 0.41 of a percentage point of the inspector general's number. The price did not require bad advice to be explained; it required a lower discount rate, which the city had and the buyer did not.

Why use 5 and 8 per cent as the two discount rates?

Five per cent approximates what a tax-exempt municipal borrower pays on long-dated debt and eight per cent what a taxable private infrastructure buyer requires on the same cash flow. They are the two ends of the ownership question, not a market forecast. Both are modelling choices and are disclosed as such: the point of the exercise is that a 300-basis-point difference in the cost of capital removes 46.70 per cent of the value of an identical stream. Test it with any pair of rates you prefer — the direction holds and the magnitude scales with the spread.

Why does the gap get worse the longer the lease?

Because the discount rate compounds against duration. At 50 years the buyer's price is 60.15 per cent of municipal value, a 39.85 per cent shortfall; at 75 years it falls to 53.30 per cent, a 46.70 per cent shortfall. Expressed the other way, extending a concession from 50 to 75 years adds 19.37 per cent to its value at a 5 per cent discount rate and only 5.77 per cent at 8. Lengthening a lease is therefore the worst available response to a low headline receipt — it sells more of the only thing that is getting cheaper for the buyer.

Can a council today achieve the inspector general's valuation?

Not by selling, unless the buyer can fund at the municipal rate — which the tax rules on private activity generally prevent. The market tied to these contracts was USD 119,846.4 million in 2025 and reaches USD 244,115.1 million in 2033 at 9.30 per cent, on bays growing 4.08 per cent, with capital per bay up 48.01 per cent. A council that wants the tax-exempt valuation has to keep the cash flow: the difference between the two prices is the value of the exemption, and it is not transferable.