Against Land Value Capture
An otherwise-good video by the Joint Transit Association about A Better Billion criticizes us for not proposing value capture to fund the scheme. I’ve seen other otherwise-good American transit advocates back this, and I’ve seen many a thinktank propose it and similar nonconventional schemes to fund public transit, in lieu of taxes. Taxes are political. Taxes annoy voters. So why not get around them by taxing development behind the scenes? It’s attractive on the surface, but in truth, broad taxes are the only way to fund government and expect it to perform as expected; value capture is so opaque that it is very easily wasted, to the point that 100% of the funds it provides can sometimes be wasted on excess construction costs, as has been the case in Hong Kong. Good transit advocates should reject this scheme and demand that funding be as straightforward as possible, with the understanding that the part about taxes that annoys voters is what ensures the money is spent well.
What is value capture?
Value capture is the name for any of a set of programs aiming to fund infrastructure by taxing the development that it would unlock – in other words, to capture some of the value gained by the private sector. This contrasts with broad-based taxes, which capture value from the entire economy and not just from specific developments or developments in specific areas. The idea is that the infrastructure generates value not just for the broad economy but also specifically in the area it serves, so it’s right to tax that area.
In practice, value capture schemes are most common when there’s perception that raising broad taxes is too difficult politically, or undesirable otherwise. Hong Kong extensively uses value capture to fund MTR expansion, not because its taxes are too high but because it wishes to keep its taxes very low. American cities have begun looking into such schemes in much higher-tax environments, with limited willingness to fund things out of the general budget; the 7 extension in New York was built with bonds tied to tax increment financing (TIF), which promised that the higher property taxes generated by Hudson Yards development would pay the bonds back.
Which projects are funded by value capture?
Naturally, value capture and TIF systems favor projects that have the most real estate value to capture. In New York, this meant the 7 extension but not Second Avenue Subway, which the real estate advisors to Bloomberg denigrated on the grounds that the Upper East Side was already developed.
This already creates biases, in favor of not just wealthier areas (the Upper East Side is after all rather rich) but also ones with high redevelopment potential, for example because they are underbuilt. This, in turn, favors worse projects, because they serve lower preexisting density.
The dominant benefit of public transportation in benefit-cost analyses, which are not undertaken in the United States, is the benefit to passengers, representing social surplus. For an example that was just sent to me, a study from last month analyzing a further Nuremberg U-Bahn extension classifies the total benefits of the chosen alternative on p. 30 as 5.73 million €/year in passenger benefits, plus about 3 million €/year in various externalities of which the biggest are reduced accident costs and reduced traffic congestion. Similarly, Börjesson et al’s ex post analysis of the T-bana, finding a benefit-cost ratio of 6, has consumer surplus dominating the benefits of the system. Land use changes are helpful, but the main purpose of a subway is to be ridden rather than to stimulate real estate development. A Better Billion looks at the possibilities of development unlocked by new lines, but not for nothing, we also look at existing bus ridership and subway capacity problems in analyzing which projects to recommend; development-oriented transit can work but only as a secondary option, and value capture overemphasizes it in preference to other needs.
Value capture and costs
Hong Kong is famous in the core English-speaking world for linking development with MTR construction. It’s also a very good example of what not to do. I complained mightily about value capture in 2017, but if anything I pulled punches, because while I did talk about Hong Kong’s use of MTR value capture as a corrupt slush fund, I didn’t know enough to connect this with Hong Kong’s other problems:
- Very high infrastructure construction costs. They are so high that value capture only covers about half the project costs, and the other half, funded directly by the government, is still more per kilometer than the world average cost. Hong Kong, moreover, has been ground zero for the adoption of the consultant-centric globalized system of procurement – British consultants who were there in the 1980s and 90s workshopped the system and then brought it home, leading to a cost explosion that rendered first London, then Australian and Canadian cities, incapable of building urban rail. In this sense, value capture is an attempt to paper over the inability to build by using a tool perfected in the city that has extreme construction costs and can only build anything because, with car ownership suppressed by taxes, it has atypically high demand.
- Overcrowding, likely the worst in the developed world. The data out of Hong Kong isn’t quite comparable to the most overcrowded cities in the democratic world (Paris averages 31 m^2/capita). Hong Kong reports median rather than mean housing size: its median is 16 m^2, and while it has a lot of inequality, it doesn’t produce nearly a factor of 2 of a mean-median spread. Hong Kong has atypically high inequality, but even that only produces a mean-to-median income ratio of 0.67, and housing surface area is distributed more equally than income. So it’s almost certain Hong Kong is the first world’s overcrowding capital, all because the state doesn’t develop enough land or housing (net annual housing growth is 3.8 dwellings/1,000 people, low for East Asia) in order to create more profits for the MTR to fund ever growing construction costs.
The way forward
Value capture and other nonconventional funding strategies should be categorically rejected. They lead to poor project selection, high construction costs through opacity, and, in the most extreme cases, other governance problems including Hong Kong’s legendarily bad overcrowding. The only legitimate way to fund public transportation and other infrastructure project is through broad-based taxes, either directly as in some dedicated payroll and sales taxes found in both the United States and Europe or, better yet, through the general budget, debated at the highest responsible level of government (municipal, provincial, or national), itself funded largely by broad taxes.
Pedestrian Observations 
Just so that we’re all on the same page without room for miscommunication, your objection is only to installing land value taxes as a direct payment method to fund transit capital expansion projects and not to a broad-based general land value tax separate from but in conjunction with extant property taxes designed to discourage underdeveloped land usage e.g. parking lots, is that correct?
Very much endorsed. The snarky Georgist version (is there any other kind of Georgist?) is that if the general fund is sufficiently depleted to make value capture necessary then your property taxes are too low, and that if public investment generates enough landlord surplus to make value capture possible then your property taxes are also too low.
Dear Alon, I read your articles frequently and appreciate your ideas, analysis, and insights.
Could you please consider writing something explaining how the consultants in the Anglophone world have caused transit costs to skyrocket? I understand the outcome in principal, but I’m curious of the mechanics and history.
Thank you for considering.
The executive summary here goes over the procurement problems. It’s not diachronic, but Borners is writing about this as we speak in the context of London.
Robert Moses used a similar scheme to fund his parkways back in the 1930’s. NYC had, and still has, a borrowing limit that’s based on the assessed valuation of its real estate. Moses was officially building parks, which legally increased property values. The parkways were legally access roads to these parks. That’s how the Belt Parkway system was financed.
Interesting. I don’t remember this particular story in Power Broker – I can’t open my copy right now to check – but if I recall correctly, the other parkways went outside city limits and were funded through legislative appropriations, with a variety of gifts, lies, unlawful seizures, political skullduggery and horse-trading, Power Brokering, and other arcane* maneuvers. And other highways and water crossings relied again on the preceding and added a combination of federal largesse, especially post-1933/34, and payments or property in lieu of taxes.
This predated his use of public authorities using toll money to repay current and capitalize future debt.
Which is all to say, the social contract implied by Alon in their post today
was violated swiftly and often. I had a military history professor back in undergrad who said “Representative democracies are a way for the state to gain access to the savings of society.” The counterpart of that is that society tends to want to have a say in how the state spends those savings.
* Thank you for providing the opportunity to use the word arcane!
The Department of Transportation seems to think they are.
https://www.transportation.gov/sites/dot.gov/files/2025-12/Benefit%20Cost%20Analysis%20Guidance%202026%20Update%20%28Final%29.pdf
You really have some kind of blind vendetta on this subject and on Hong Kong’s MTRC. In this new piece you are attributing high construction costs with Land Value Capture (LVC) but does anyone believe those costs would be any lower without LVC? They are only related in the general sense that everything in HK is determined by the upper echelons of the corporate elite (and of course these days, Beijing).
Except that it has built one of the world’s best Metro systems, and gets voted World’s Best in various polls. And as I suggested last time (see below) in some ways it is an anti-corruption scheme because the money can go nowhere except to transit. It is true that, as I suggested and you are repeating here, it is a partial substitute for direct taxes. But what is wrong with that when the likelihood of getting such taxes to fund public infrastructure is so hard to obtain? And yes, that resistance by top earners in HK (and everywhere in the world) is a manifestation of corruption or at least undue influence but the solution is not corrupt but simply a rather elegant rational workable one.
Anyway, rather than repeat myself afresh I will repost my entire post from your earlier piece for your reader’s delectation (without direct links to try to avoid the spam filter): (I have added some emphasis).
[pedestrianobservations.com/2017/09/07/meme-weeding-land-value-capture/#comment-25783]
Michael James 2017/09/07 – 04:55
Alon, this is an extremely disappointing piece. I can only conclude that you must be politically/ideologically driven to make so many incorrect statements, especially as I have specifically addressed most of them on your blogsite in the recent past. I think it is outrageous to call what happens with HK-MTRC as corruption. It’s what I call good planning to bring stable finances to the crucial transit system for the overall good of the city-state. That is why several Chinese cities (Shenzen …) have adopted the same approach and other transit specialists commend it (more in a separate post).
This is very misleading. Even Stephen J. Smith calls it “a public-private company”, because that is what it is. Still 70% owned, and obviously largely controlled, by the HK government. The minority of shareholders is presumably to bring the much-vaunted private-sector financial discipline blah blah (though I don’t know who those investors are–are they truly non-governmental?).
and:
(That’s confusing and seemingly contradictory but never mind …) In both cases (and also French-SNCF’s TGV operations) this is merely an operational profit which is fair enough, but of course the capital cost will never be repaid. This includes all of the “privatised” JR companies who have never had to repay or buy any of their huge assets (indeed massive debts remained in a government holding entity that continues to lose money to this day). For the same reason it is dubious to consider those JR companies as “entirely private”, indeed extremely misleading IMO.You haven’t given any concrete or meaningful example of corruption in the HK case. Stephen J. Smith does not use the word in his piece even though he too doesn’t seem to approve of the arrangements but then his history involves the Real Estate industry, and to be sure there are property developers in HK (many billionaires) who object to the fact that they are effectively cut out of some part of the development that MTRC does. But the reality is that private developers have plenty of opportunities in all the developments that MTRC [do], and I suspect any whinging is limited (they surely must realize these developments wouldn’t happen at all without the MTRC).
This is your what you consider corrupt:
Now, if this profit was going back to shareholders or some other private entity you might be right. BUT IT ISN’T! It is designed so that the public-private entity is financially viable and can also raise some of the huge capital costs involved in expanding the transit system (more later). Also there is nothing the least “surreptitious” about it. It would be seriously pointless just the government selling the land or development rights directly to the developers because that money would be a one-off infusion of cash that would disappear into general government revenue. Instead, MTRC has a permanent income stream from its developments that the politicians can’t get their miserable pork-barrelling hands on. If anything this is anti-corruption. Further I reckon it creates better type of development than private developers left to themselves (this would be even worse, is even worse, in the west or at least the Anglosphere).
Doesn’t something similar happen with US airports (essentially independently-operated but ultimately owned by various governments, federal, state or local): do the airport companies sell off their retail portions to a developer in a one-off deal, or does the airport recover rents from those retailers directly? I know which makes the most sense (even as developers probably scream about it: don’t tell me, I’ll bet there is a law against this arrangement in Texas ….).
No, he didn’t. There is no use of that word and really not even the accusation or implication. You’ve got this seriously wrong and I ask myself why are you doing it. The citation from the South China Morning Post is faintly ridiculous and of course merely reflects some actually corrupt business interests; I wonder if Murdoch was still the owner of the SCMP when that was published? And does anyone doubt that a rail-line and big station with associated development doesn’t, almost “magically”, create value? It’s just sour-grapes that those developers aren’t the sole beneficiaries.Stephen J. Smith writes:
That is another misleading simplification. Building costs are so high that MTRC contributes about 50% of the capital cost of any new line extension and stations, ie. the government must still find about 50%; but of course under these value-capture arrangements MTRC will recover more over a longer timescale from those developments. It’s win-win for the city-state of Hong Kong and I have no idea of why you want to deny this.
This is simply untrue. Hong Kong has less buildable land than almost any city you could think of. And it rather belies that point you are failing to make, that MTRC’s most notable value-capture developments have been on reclaimed land, such as Kowloon-West, the massive IFC scheme in Central and the Tung Chung development opposite the airport (which, even if on the biggest island in HK, Lantau, the vast majority of it being undeveloped forest on rocky mountains & hills, nevertheless at least half of the land of Tung Chung was reclaimed from the sea–for the obvious reason that the mountains of most of Lantau go right to the sea). Further, MTRC has only a minority share of all development at Tung Chung.Of course exactly this has happened in NYC, in the reclamation by the city (NYNJPA) to build the original twin towers WTC then Battery Park City adjacent; the NYNJPA acts as the master developer in all these things, yes? Though I have asked here recently whether the MTA recovers any direct and recurrent income via rents etc of the vast shopping precinct built above its stations (run by Westfield, a rapacious Australian property developer)?Grumbling about the “low quality” of Hudson Yards sounds like a bit of corruption in itself; I mean does anyone doubt it will turn into highly valuable real estate with time? Then there are “air rights” which presumably apply to any redevelopment of, say Penn Station, etc. These should belong to the MTA or at least shared with the city so that a permanent source of income offsets the high capital costs and operational costs. I believe the same operates at Chatelet-les-Halles in Paris, the world’s biggest Metro station with a very large retail complex above and fully integrated.
The efficiency or inefficiency of how different transit authorities operate is really a different issue and should not be wrapped up in the argument.
That only deserves one response: nonsense.
I think that is actually a pretty good question that might be answered by looking at same city examples. Does the price of rapid transit construction in Tokyo and Osaka vary significantly based on how reliant they are on land readjustments or redevelopment of former industrial sites for funding? The Minatomirai Line was very expensive after all.
Japan barely has land value capture. It just has high property taxes, so high people often reject inheriting land at least in rural areas. Land Readjustment isn’t really LVC so much as a site-specific land assembly tool, to get around Japan’s weak and hard to use eminent domain powers. And for city centre subways its generally not used because its too difficult. The lines that used most aggressively are suburban extensions, e.g. Tsukuba Express or Tokyu Denentoshi.
Japanese private rail’s non-rail businesses are a kind of value capture sure, but its a bit like a private equivalent to conventional broad based taxation. More riders and shoppers are more important to them than maxing out station land values. Yes it makes them much more willing to contribute 10-30% of costs for stuff like the Osaka Namba line or the East Kanagawa lines.
I’ve uploaded all Yokohama subway line costs to the TCP database, Minato-Mirai line is very much an exception. It has the most architectural bloat of any underground line in Japan, both in finishings and sub-ground space taken. Additionally soil is waterlogged mess and it crosses street grids multiple times, which means it had to pay easements.
And the just opened Chuo line Yumeshima extension for the reclaimed island which had last year’s world expo seems to have been built very cheaply both because of decent project management. Although I fear it will fit into Kansai bad history of reclaimed land boondoggles. At least it isn’t as expensive as Kobe’s Kaigan line i.e. the worst subway line in Japan.
That’s not true for the Shinkansen lines at all, sans the paid off Tokaido, JR East and JR West are still paying the construction debt for Sanyo and Joetsu/Tohoku Shinkansen. And most of those debts were not incurred building useful things of the JNR era like quadtracks or the Musashino line but boondoggle lines in rural area and inefficient labour practices. And those were costs chosen by the politicians at the time, not the private companies that existed afterwards.
And the main reason why the holding company made losses on JNR debts was because nobody in 1987 was planning on commercial land property prices to fall 80% once the bubble burst . And JNR privatisation doesn’t have much to do with structural macro-policy, which has consistently screwed the Japan rail industry I would argue.
Private vs Public in any system is always to some extent an abstraction since private property rights require a public to recognise them. Japan is always a bit explicit in its corporatism, but its always there in any system.
JR’s show much more operational autonomy and willingness to make ruthless trade-offs than JNR did. You can see that in JR Central’s screwing Shizuoka to max nozomis between Kansai and Tokyo or JR East’s recent fight with eastern Chiba over its turning express services into the local to help increase frequency at inner suburban stations. And the worst run JR, Shikoku is visibly the one with the least autonomy and most politicised operations.
Isn’t the Minatomirai Line being an aberration in support of land value capture providing bad incentives? Would it have had as much architectural bloat if it wasn’t built as part of a massive redevelopment of former industrial land into a new CBD zone?
There was zero new track for Takanawa Gateway, but it seems believable that if there was, the cost per km would be egregious as well. I guess the Yumeshima Extension is a counterexample where they had more discipline. However the potential land use intensity, even with the Expo plans, was never going to be comparable to redevelopment of city center industrial land. That applies to other more suburban lines as well.
Something like that. And the core concept of Minato-Mirai has worked, a new extension of CBD redeveloping the old port district. Overall its probably the most successful of its type in Japan. I can’t speak to the exact details of how blew out of proportion.
And actually if you look at Portland redevelopment schema across urban Japan, mostly the transport is cheap, Portliner/Rokko systems in Kobe, Yurkiagome in Odaiba Tokyo, New Tram in Osaka, BRT in Fukuoka’s Island City, Aonami line in Nagoya* etc. They are often cheap, but under perform because the underlying project was based on false assumptions about price and where people wanted to live/work. Minato-Mirai is the opposite of those, overpriced construction in service of a good project. Like the Jubilee Line Extension in London. Which did influence everybody with its “world-class” (derogatory) architecture.
Takanawa Gateway is a different since JR East is self-funding the thing and its a part of long redevelopment at a time when the central Tokyo high-rise condo and office market is red-hot. Its deeply integrated into a wider real-estate project, so JR East is paying for something to get something. Additionally the site was difficult they did have to careful building around busiest rail corridor in the world after all. Plus Kengo Kuma isn’t cheap ( I have mixed feelings about him, his recent megaprojects are meh, but I loooove the work he did for the Nezu museum).
*I will add some of these to the database soon. I have the numbers from enough to confirm pretty good costs but breaking them down to the standards of the database (I am the most anal uploader on the team).
Property tax, it’s called property tax. Almost everywhere it’s based on current-ish market value. When the value goes up because the property can put “walk to train” in the ad it gets taxed. Because the market value went up. Which is taxed.
20 billion dollars of skyscrapers get assessed more than the air that was west of 10th Ave. before they built 20 billion dollars of skyscrapers. When they do the part west of 11th Ave. it will likely be another 20 billion or so. And the sales taxes the mall under the skyscrapers will generate. Which is more than air generates.
Much more than the air that was between 10th Ave. and the river before Hudson Yards was developed. Air doesn’t get assessed.
Adirondacker is quite correct but the problem is how do you get general property tax like that spent on transit. That’s what the HK model is all about. It clearly works. Alon wants to complain that 50% of it is wasted on inflated cost of construction but many places have very high cost of construction without such LVC going to transit. The obvious thing about places like HK, London and NYC etc is that they are rich cities and this is a method to squeeze out some of the immense value held in property in these places to fund public infrastructure, and, critically, fund it sustainably not just in one-off grants at the whim of politicians, local and national. In places like Paris and France you may not need it because governments of all shades support public transit (remember that GPX, Europe’s largest civil construction a.t.m., was begun by Sarkozy), though in some ways the versement transport tax substitutes; however the VT is a payroll tax and arguably deters employment and unlike a LVC does not capture the huge increase in land value that is typically way above inflation in these cities (probably not in France, another story). Indeed I would say that the LVC is a tax on assets which is exactly what is required to rebalance inequalities (a bit).
Incidentally the purely private version of this in Tokyo is IMO worse because it doesn’t produce the best transit (despite the hyperbole by non-Japanese who don’t commute in Tokyo) especially underinvesting in “mature” systems because the original land value has been captured by the same company who doesn’t have motivation to keep investing in the transit (instead “let them squeeze up” to impossible densities). Plus transit fares are high; compare with the low fares and better quality travel experience in HK (and Paris).
Versement Transport financed most of the recent urban transit investment but does not contribute to the Grand Paris Express, and Land Value Capture is not envisioned as such.
The bulk of the Societe du Grand Paris’ resources comes from taxes on offices, warehouses, commercial and parking spaces. A tax on RATP’s rail rolling stock covers about 10% of its budget. About 2.5% of the budget comes from a 15% share of the hotel sojourn tax. The French State can be creative and new taxes may come into play before the SGP’s debt is extinguished decades from now.
Governments of all shades supported the GPX but not with cash. On the contrary, the Societe du Grand Paris was asked to finance projects beyond the GPX (extension of Paris metro M11 and M14, completion of RER E). The government should benefit from property taxes on the development around GPX stations but GPX operation losses will have to be covered.
Right, but as I wrote, the capital requirements for these major infrastructure projects is huge so a LVC can only do part of the job, about 50% in HK. The important point is that it keeps contributing year after year and grows as TOD grows and becomes more valuable. Oh, and the funds go directly to transit and cannot be easily changed by transient toxic politicians or their lobbyist masters.
I can only guess that France doesn’t use LVC due to political perceptions. And that the VT is a payroll tax that is more directly justified because it is the workers (and their employer) who benefit from the transit.
By electing responsible adults. Unfortunately voters elect people who tell them the Magic Laffer Curve will fertilize the money trees so well we will all be rich enough to drive everywhere.
This. If you have value capture systems, it means your conventional taxes aren’t doing their job. Remember its not just Property taxes, but your capital gains, income, corporation taxes if they are working should also be hoverring up the value uplift of building permissions. Land rents have to be cashed in too.
Land value capture is a nimby system to tax new build instead of existing property.
Hi Alon,
Thank you for this post.
I’m curious of your thoughts on the spending side corollary: i.e. cost shares between the responsible levels of government and the next highest level. I’m thinking specifically of USDOT grants to states and cities either for operations or capital projects. You’ve written in the past about the Other People’s Money problem, and the Baboon Theory (loved that) – do successful transit systems in Europe and Asia resist this by keeping both taxation and spending constrained? For instance do German or Swiss S-bahns get federal funding in addition to state/canton/city? I’ve long taken for granted that American cities can never pay for large transportation projects themselves but I don’t know how the rest of the world works. If I remember – it was a post some time back – you might have mentioned that Switzerland’s integrated system with a focus on mode integration, scheduling and efficiency over speed came about in an environment of scarcity, after voters rejected more ambitious and extensive plans.
I guess what I’m really asking is: Do spending constraints and scarcity breed innovation and efficiency?
Capital construction projects here get extensive federal funding, yes, at 50% or even 75%. Likewise, in Sweden, non-capital cities compete for state grants for public transport projects. (Stockholm decisions are made directly by the state due to the size of the projects there.)
Spending constraints by themselves are fairly neutral. It’s much more important to focus on transparency at all levels, including transparency of item costs, broad taxes, and clear promises that the public can judge the state against (so, no state of good repair scamming).
Thank you for replying!
About transparency in particular: It is extremely frustrating when contracting authorities are not public with their data. The cost to them is incremental. It’s often, if not always, required by law. They just don’t want to share.
Here in Connecticut, the state carried out an electrification feasibility study of the New Haven-Springfield Line, Danbury Branch and Waterbury Branch. The formal study report, if there was one, is not public. The top line results were buried in a briefing (2-3 slides worth) to the CT Public Transportation Council, in theory a collection of elected officials who represent mass transit riders and advocate for services but in practice a passive recipient of PowerPoints from ConnDOT. The cost results were so absurdly high ($42M/mile, 2040$), with no data on the assumptions and categories therein. I can’t tell if it’s garden variety incompetence or intentional sandbag.