I'm pleased that something is being done about this. rte.ie/news/politics/2026/09…

The Metrolink from Swords is, in a way, like reclaiming land at Sandymount Strand. In both cases you’re spending money up front to create new capacity to live within a 30 minute commute of jobs in central Dublin. Yet our appraisal process doesn't formally recognise any of the value created by additional housing. progressireland.substack.com…

Too few trains to justify the homes; too few homes to justify the trains. Copenhagen did both at once: built the metro, zoned and developed public land along it. Two sins of transport planning 👇 progressireland.substack.com…

California's sin: build heavy rail through suburbs that don't need it, and never zone homes around the stations. A textbook white elephant.

Last year I rode a big, modern, and completely empty train to Berryessa in California. Copenhagen's metro, by contrast, carries 50x the riders per dollar spent on the California line. A lesson on how to assess rail projects. Thread below. 🧵 Link: surl.li/xubnmx

If inference is the labs' product, this metric fairly represents gross margins

This street in north Dublin. Lost: six parking spaces. Gained: a nice new vibe which will be reflected in higher house prices. In Toronto, each (mature) street tree added €100k to aggregate local property prices. Six smallish trees here — that’s €100-500k in extra value

Big metro stations are very expensive, said @EnglishRail. Also: modern automated trains accelerate and decelerate faster. So: smaller platforms with more frequent stops can move the same number of people at lower construction cost. This was Metrolink's design approach.

4. Since the 2021 appraisal: costs up, population up faster than forecast — and the standard method counts zero land-use benefits, the main purpose of an urban metro. When Sydney measured them, land use was 20% of a line's total benefits.

3. Airports are not normal places. More air connectivity means more HQs, more service jobs, faster local growth. An airport anchors a metro line with all-day demand in both directions.

Next Wednesday, cabinet decides whether Metrolink advances to final tenders. Here are four good reasons to approve it. 1. Per kilometre, Metrolink would be mid-table among peers.

Is it a coincidence that Ireland has the lowest share of electrified rail in the EU?

Irish transport cost-benefit analyses assume a project won't change how nearby land gets used. This discriminates against the most transformative projects, since hanging the city is the entire point of a metro.

Nice new neighbourhood in... Nairobi unityhomes.co.ke/silver-hill…

It has been drawn to my attention that this is missing a picture

The teens talk of little else

Irish people would be much much wealthier if they owned more stocks. Tax makes it not worth our while. Chart by CGPT

I’m in France. The state is present and unabashed in a way that this Anglophone finds strange. Pics: local post office, “we honour the elected”, exquisite public realm I and II

“Our diagnosis of Ireland is that it’s bad at building. A right-wing solution to that might be, ‘Deregulate the hell out of everything.’ The opposite side might say, ‘The problem is too much market.’ Our agenda borrows a bit of both.” The full Indo profile of progress Ireland: sl1nk.com/98dl48k

1500 units per hectare 😮 nytimes.com/2026/06/15/opini…

If DPER loves Metrolink, it should let it go Annual cash control is a poor substitute for project governance. Metrolink's railway order has been granted and procurement is underway. Yet there are still important decisions to be made. One decision is over who, exactly, will control the project. Who will make the big decisions, what experience will they have, who may they afford to hire, how much will they be paid, what decisions will be theirs to make, and under what oversight will they operate? One of the most important questions relates to the terms on which money moves from the exchequer to the project. The control of funding is a form of control over the project. What's the right way to think about funding? Too little oversight In life, there are two ways to screw up: by doing too much, or too little. The funding of metro construction projects is no different. The people who control the purse strings can screw up a metro project both by funding it too loosely, and too tightly. The first of the two failure modes is straightforward. Obviously, it's a mistake to carelessly fire money at a metro project. If the funders don't care how money is spent, the project managers and engineers won't care either. The best examples of too-loose funding controls come from the US. In the US, funding for infrastructure can come from many pots: city, state or federal. This gives rise to a situation where a city or state-managed project could be largely funded by the federal government. In Boston's "Big Dig" tunnel project, for example, the state of Massachusetts was responsible for delivery, but the project was scoped and designed on the assumption that Washington would pay up to 90 cents of every dollar, as it had under the Interstate highway programme. The crucial spending decisions were made by people who believed the bill was going elsewhere. By the time the federal contribution was capped at roughly $7 billion, the damage was done: construction costs had risen from an initial estimate of $2.8 billion to $14.8 billion, or over $24 billion once the interest on the borrowing is counted. One would imagine the best way to avoid a Big Dig-style catastrophe, then, is to keep a very close eye on spending. But in practice, things are not so simple. As many projects go wrong due to tight budgeting as loose budgeting. Too much oversight What does tight budgeting look like in practice? It usually means the project needs to go back to the treasury every year for fresh funding. Or that the project leader is constrained in the spending decisions they're allowed to make. The first problem with annual funding is that it breaks a big, efficient thing down into smaller, less efficient chunks. A project whose funding renews annually will be continually accelerating and decelerating. Its crews and machinery will power down and power back up, losing momentum in the process. Its work will be sequenced sub-optimally. Its learning curves will be slower. And it will simply take longer, which is in itself more costly. This last point matters a lot. The megaproject scholar Bent Flyvbjerg advises to "think slow, act fast": once construction starts, every additional year of delivery is another year exposed to recessions, elections, inflation shocks and scope creep. He calls it the window of risk. Annual funding mechanically lengthens delivery, the exposure window, and expected costs. The other problem with annual funding is that it makes it harder to have trusting, reciprocal relationships. Mega-projects are a collaboration between organisations with different interests. In the good scenario, they trust each other and work towards a shared goal and everybody wins. In the bad scenario, everyone retreats to their corner and gets the lawyers in and you end up with a National Children's Hospital. How does tight control over spending lessen trust? One example is risk: if contractors perceive a risk that their machinery will be sitting idle as a result of spending interruptions, they will charge more up front. Another is competition: the perception of a long term pipeline of work will attract competition, which will discipline everyone. Another is that stable long-term investment in the project justifies the investment in in-house staff, which are associated with lower costs. Another is that tight control over spending gives the project less leverage in its negotiations with suppliers. What's the evidence? What's the evidence for these big claims? The three best sources are the Stewart Review, which digs deeply into the failings of one specific project, HS2; the Lovegrove Review, which teases out the implications of HS2 for civil service processes; and the Transit Costs Project, which looks broadly at what drives metro construction costs globally. The Stewart Review said: "it is almost impossible to manage such a large project on annual funding settlements alone." "The absence of a multi-year spending review moved the Programme to a position of annual funding settlements with no long-term funding control period in operation... This all led to scope deferments, cancellations and supply chain uncertainty. It also undermined HS2 Ltd's negotiating position with suppliers," it added. Stewart's recommendations: a dedicated budget line for the programme, funding controls extended across five-year periods, and the flexibility to move money between years. The British government accepted all nine recommendations and establishing a multi-year settlement of over £25 billion for HS2 in its 2025 Spending Review. The UK government also commissioned Sir Stephen Lovegrove to review the lessons of HS2 for how government oversees its delivery bodies. His review said "Managing Public Money rules are better suited to the assessment of fixed-point decisions or projects rather than the multi-decadal and multi-phase reality of the very largest programmes." The Lovegrove review said the refusal to hire experienced staff into the civil service was a "false economy". "If the corollary of this is that significant pay freedoms need to be in place in order to attract the volume and seniority of talent to full time Civil Service positions, they should be given," it added. Similarly in the US, the Transit Costs Project found that agencies "take advantage of federal grants and transportation bond acts… to pay for consultants to do jobs that previously would have been done by in-house engineers." The need for funding at the scale of the project is a theme of the Transit Costs Project. It said, of Naples' metro system, "the stop and go nature of funding and construction has contributed to cost increases." Of Rome's metro, it said, "the choice to break down a larger metro project into smaller, more financially palatable sections resulted in higher overall costs at the end." Of New York, it said "a lack of leadership and funding certainty at the state, local, and agency level enables [New York's] costs to outstrip those found in Istanbul, Italy, and Stockholm." Of New York: "uncertainty over funding can lead to tendering delays, which inevitably, at its most benign, lead to inflation-driven cost escalation and schedule delay." The Transit Costs Project found, of Italian projects, that "Cutting projects into shorter sections to make them more financially palatable in the short run ends up increasing the overall costs in at least four different ways: First, by making economies of scale more difficult to achieve; second, by reproducing expensive construction staging and operational requirements; third, by hindering the learning curve of both management and contractors that is typical of all large-scale complex projects like metro construction, and, fourth, by introducing potential delays that slow down approvals and trigger cost escalations." The cleanest natural experiment comes from the US Navy. Congress normally funds shipbuilding one year at a time, but certain multiyear procurement authorities allow the Navy to commit to several ships at once. The Congressional Budget Office and the Government Accountability Office have repeatedly found that multiyear contracting saves in the region of 5 to 15 per cent per ship through the mechanisms described above: stable supplier workloads, bulk material purchases, retained workforces, faster learning curves. The converse has also been measured: the Pentagon estimates that operating under annual funding uncertainty costs it billions of dollars a year. The experts say: With Metrolink's governance, the state needs to thread a needle. Not so little oversight that Metrolink spends carelessly; not so much that it can't operate freely and set the right expectations for stakeholders. One caveat: stable funding is necessary, not sufficient. London's Crossrail had a protected, ring-fenced funding envelope and still overran by around £4 billion, because its governance and its grip on programme integration failed. A multi-year settlement must come paired with the capability and oversight reforms that Stewart and Lovegrove describe. But without the funding settlement, none of the rest is achievable. Ireland's good fortune is that we're acting after peer countries, so we can learn from their mistakes. The Stewart Review, the Transit Costs Project, the Lovegrove Review are full of recommendations. They emphasise: - Agree a multi-year funding control period, not annual settlements alone. - Allow the delivery body to commit funds across years where contracts require it. - Delegate defined authority over contracting and sequencing. - Permit pay flexibility to hire senior commercial, engineering and project-controls talent. - Maintain strong sponsor oversight through monthly reporting, independent assurance, cost/schedule baselines and reset gates. The stakes are high. Metrolink could cost anything from €9.5 billion to €24.7 billion. As Bent Flyvbjerg argues, megaprojects do not go wrong; they start wrong.

Come along!

I’m in the Irish Times today making the case for an open mind on SMRs. Like renewable power in the 1990s, they have potential

Too much chaos in a city plan — bad, messy, inefficient. Too much order — cold, inert, lifeless. There is a sweet spot.

A fine meeting this morning with The Confederation of Swedish Enterprise. The Swedes are interested in Irelands outward-facing growth model. And Ireland has much to learn from Europe’s most entrepreneurial country. @SODaunfeldt @svenaringsliv

I'm in the IT today talking about the nuclear ban. irishtimes.com/environment/c…

The public luxury of the Iveagh Gardens

YIMBYs and the left: much in common.

I’m in the FT, talking about Gerry Hutch

Quango formation in Ireland seems to have mainly been a Bertie-era FF thing

Nice new paving on South Anne St, Dublin

From @aarmlovi, a reminder that it’s a mistake to fixate on land prices. High land prices can coexist with low housing costs provided you use the land intensively. It’s when you cap density that land prices became a problem.

Quoted in the Journal.ie about the procedure fetish

I asked CGPT for an Irish one; CGPT is abundancepilled

Small modular reactors could complement Irelands wind energy system beautifully.

We got a sweet new office on Dawson Street!

At the @ProgressIreland monthly drinks, @o_mcpartlin talked about the multifaceted success of the city of Madrid

I had the honour of speaking on a panel at the Assistant Secretary Network annual conference last week. It was great to meet civil servants in their element. Day and night, all the chat was about how to improve the workings of government.

extra.ie/2026/04/26/news/met… Some of my remarks to the Committee on Enterprise, Tourism and Employment got picked up this morning. On the decision about how to govern and staff Metrolink, a lottttt of money is at stake.

Why are modular homes in gardens good for the system? Offered my thoughts to the Sunday Times:

What I'll tell an Oireachtas Joint Committee on the topic of Competitiveness and the Cost of Doing Business later today. Part one: Progress Ireland Chair, members of the committee, thank you for the invitation. My name is Seán Keyes. I am the executive director of the independent, non-partisan think tank Progress Ireland. I’d like to start by explaining who we are, how we work, how we’re funded, and what we’re trying to achieve. We were founded in 2023. Our board is made up of Fiona Cormican, former CEO of Cluid Housing, Jim Breslin, a former Secretary General; Donal de Buitléir chair of An Gardaí Síochána; and Brigid Laffan, Chancellor of the University of Limerick. We work on housing, infrastructure and innovation policy. Our modus operandi is to connect Ireland to policy solutions with a proven track record from overseas. How do we choose which policies to focus on? Policy choices are ultimately determined by our values. We believe in building as a tool for solving policy problems. We think Ireland’s most pressing problems – from competitiveness to the cost of living, to climate, inequality and polarisation – will be solved only by building. Ireland needs to build new housing, infrastructure, companies and institutions. We believe state capacity matters. We believe the state should be capable of following through on its priorities, whatever they might be. A government should be capable of doing the things it wants to do. In an increasingly complex world this is often not easy. We believe in reforming the state so it can follow through better on its goals. We believe in progress. We believe the world can be much better than it is, and Ireland should aspire to be much better than it is. Ireland could have the highest living standards in the world. The next point is about funding. We are often asked who funds us, and what influence our funders have over our policies. These are legitimate questions. We are funded by private philanthropy. Our funders support us because they share our values and believe in our policies. Funders don’t have influence over the policies we work on. They do not benefit directly from our policy work. And we publicly disclose all material funders. Part two: Our diagnosis The committee is concerned with competitiveness and the cost of doing business. Ireland is a country beset by shortages of physical things like housing, power, water, trains, and ports. Ordinary people experience these shortages as high rents and costs of living. Business owners experience them as the cost of doing business and foregone investment. They are two sides of the same coin. Investment, on both big and small scales, is bottlenecked by our housing and infrastructure. Citing lack of housing, Global investment banks choose Amsterdam over Dublin for their post-Brexit EU headquarters. And small enterprises struggle to pay their workers a wage sufficient to cover their rent. Progress Ireland’s high-level diagnosis of Ireland’s situation is that there is extraordinary demand to live and invest in Ireland, and the state is struggling to build enough to keep up. Here are four charts setting out that case. The second point we would make is that in Ireland, money is usually not the bottleneck. Despite plenty of money, projects are slow to get started and finish. Progress Ireland sees its role as identifying and helping remove bottlenecks. Part three: Our solutions I will briefly describe four policies we are working on. Each one has the potential to have a big impact on the cost of doing business; each has a strong track record; and each is practical and implementable. Land readjustment. What does Ireland need? It needs hundreds of thousands of homes near jobs; in nicely masterplanned walkable communities; without emitting lots of carbon; without bankrupting the exchequer. Land readjustment is a tool designed to solve this problem. It is used to masterplan neighbourhoods at scale so that they are self-funding. It is used in more than 30 countries including Spain, Germany and the Netherlands. 30 per cent of urban Japan was built with it. How it works: a qualified majority of landowners in a designated area to pool their land, cede roughly a third for roads, parks and services, and receive back smaller but much more valuable serviced plots. This tool shows how land readjustment creates value for landowners and pays for infrastructure. Electrified Rail. A natural complement to land readjustment is electrified rail. Rail’s superpower is its capacity: electrified rail lines can move a colossal number of people. One fully souped-up rail line can move as many people as 42 lanes of motorway. Pairing electrified rail investments with new housing can unlock enormous amounts of housing in every urban centre in Ireland. Street votes. Most Irish jobs have been added in towns and cities. It’s in these places that it’s hardest to build new housing. The result has been higher housing costs and longer commutes. Street votes is a policy designed to gently intensify housing where it’s most needed, and nearest to jobs. How does it work? It was first pioneered in South Korea, Israel and the UK. It lets the residents of a single street vote, by supermajority, to grant themselves permission to extend or add storeys to their homes, subject to a mutually agreed design code. It unlocks gentle density in a way that is consensual, local, and does not require top-down planning fights. Metrolink and other complex projects. As Ireland gets bigger and richer, the state will have to deliver bigger and more complex projects. The children’s hospital was one. Metrolink is another. The DART+ Tunnel, or better government software, might be next. To deliver complex projects quickly and cheaply, the government should employ and empower more experts. Take Metrolink. When it comes to metro construction cost per kilometre, there’s a striking difference between English-speaking countries and European ones. English speaking countries pay approximately 4x more for their metros. Why is that? European countries tend to be better clients. Their governments employ expert builders. And those experts are empowered to make decisions. When it comes to Metrolink, our research suggests a €52 million investment in expert staff inside Metrolink — hiring the engineers and project managers directly — could save up to €15 billion on MetroLink alone. The model is the NTMA. I thank you for the opportunity to speak and welcome your questions.

Delighted @ProgressIreland played a part in this. Thousands of small homes will soon be popping up all over the country. More options for renters and lower rents for everyone. And for homeowners, help with the bills.

It’s hard to get data on the quality of infrastructure. This one is derived from a survey of business managers and published by the IMF. But it’s from 2017. And when I looked into it, the same sizes were tiny. IMF doesn’t make this anymore AFAIK

Neat interactive model by @o_mcpartlin showing how housing regs flow through the amount of housing construction: progressireland.org/models/h…

The dark brown line is our current system, the light brown one is the Swedish one we're reportedly looking at moving to, and the yellow one is the British system. Moving to the Swedish system would leave savers about 1/3rd better off after 30 years.

Wow what a turnout upstairs at Doheny and Nesbitt

If these ideas are wrong, we're wrong. I sometimes find myself talking past people. The topic at hand might be rent controls or housing funding models. But the discussion will fail because my interlocutor and I disagree on some foundational idea, four steps prior. We can’t fruitfully discuss funding models for housing because we don’t even agree on whether its desirable to build housing for profit. Progress Ireland’s newsletters often skip a lot of context. It’s assumed the reader understands what we want (cheaper, nicer, better located housing for ordinary people; world class Irish science and commerce). And it’s assumed the reader understands the pillars of our worldview. But the pillars matter, and we shouldn’t always skip them over. If they’re not right, the whole edifice tumbles. When it comes to housing, there are seven of them. We think policies that take these ideas into account have the greatest chance of success. Housing 1. Building more houses drives down house prices. How much? By somewhere between 1 and 2.5 per cent per 1 per cent increase in the housing stock, per most studies. This applies to any type of new housing construction — market rate housing as well as subsidised housing. 2. Building market rate housing makes all housing cheaper. Studies by Evan Mast in the US and Cristina Bratu in Finland find new market rate housing begins a chain reaction that lowers housing costs for the entire system, including low-income tenants. 3. Housing supply is constrained by the planning system. This is obviously true to some extent, because the purpose of planning law is to control building. But we believe the effect size is big. We think, for example, that exempting seomraí from planning would result in lots more seomraí than a world where they must first get planning permission. A US metric called the Wharton Residential Land Use Regulatory Index (WRLURI) hints at the impact of planning law on housing supply. It’s an independent measure of how restrictive planning rules are in individual US cities. The following scatterplot shows the relationship between house prices and the strictness of planning rules. Places with less restrictive planning rules build more housing and as a result, house prices are lower. 4. Ambiguity around planning rules reduces housing supply over and above the restrictiveness of the rules. Ireland’s planning regime is discretionary and ambiguous, which adds extra risk to construction and reduces supply. Ireland’s system is modelled on the UK. By contrast, planning law in the rest of the developed world tends to be more specific (housing must be X distance from path, y ratio of walls to doors, z distance from neighbour…). 5. Planning law is downstream of local politics, and any enduring solution needs to take locals into account. As the experience of New Zealand under PM Ardern showed, it’s possible to railroad people into looser planning rules. But the rules won’t endure if the public hates them. Ardern was thrown out and her reforms didn’t last. Lasting solutions take the views of locals into account because they effectively have a veto – whether we like it or not. 6. High construction costs are a big drag on Irish housing supply. Irish construction costs for apartments are about a third higher than those in European peer cities on a like for like basis. This is a separate problem from Irish planning law, and has similar effects on housing supply and prices. 7. Development can’t be directed around the country by government policy. This is the conceit at the heart of Ireland’s regional development policy. The policy intentionally limits growth in Dublin, as a means of encouraging growth in the West and South. We believe these efforts make Dublin poorer and more expensive with almost no compensatory benefit elsewhere. Housing policy reforms we are inspired by: Washington DC, Sydney, Minneapolis and Austin’s city-wide upzoning (aka increasing the number of homes permitted per plot) which reduced housing costs relative to peer cities; a flurry of pro housing bills introduced in California in 2022; Houston’s use of “opt-out” mechanisms to win broad support for city wide upzoning; the UK’s street plans scheme (through parliament but not yet finalised) which lets local areas decide on their own housing rules; the recent proliferation of accessory dwelling units across the USA; Israeli reforms that gave homeowners an incentive to favour more housing; Japan’s implementation of “laddered zoning” which allows housing to built on almost any type of zoned land; and New Zealand’s plans to incentivise local authorities to support more housing. Infrastructure The three legs of the stool of efficient infrastructure delivery are: lots of skilled contractors; a competent client; and clear, fair and speedy planning regulation. Our first policy is focused on the need for better state clients. We believe the state, as the owner of infrastructure projects, has an obligation to understand projects, manage partners and bear risks. We believe this role matters more in more complex projects. We believe a state that lacks this technical capacity won’t be able to deliver complex infrastructure quickly or cost efficiently. European governments’ projects are more often managed by in-house experts. In English speaking countries, the projects are usually managed by generalist civil servants. In the top seven or so European countries, complex infrastructure projects like metros cost 20-30 per cent those of English-speaking countries. Experts in metro construction costs attribute the difference, first and foremost, to a lack of client-side expertise in English-speaking countries. The next step of our infrastructure research will be focused on how to attract the best international partners to Irish projects. We believe a deep market for contractors will result in cheaper, faster projects. We believe the best way to attract partners is to de-risk infrastructure projects as much as possible. This can be done by for example, providing a clear and certain pipeline of projects; minimising planning risk; paying bid costs for major projects; breaking large contracts down into smaller ones. The following step will be focused on new funding models for infrastructure. We believe infrastructure construction should work hand-in-glove with property development. In the countries that are most proficient at building transport infrastructure — for example, Japan, Hong Kong, Australia and Denmark — profits from property development pay for infrastructure, in a virtuous cycle. Infrastructure policies we’re inspired by: the Metropolitano Milanese (MM), a state-owned company that built Milan’s world class metro system cheaply and now exports its skills around the world; the Sinema, an entity brought together to deliver Madrid’s metro and then dissolved; Australia’s Infrastructure Commission which developed a clear pipeline of work and made Australian financiers and P3s world leaders in infrastructure; Tokyo’s and Hong Kong’s development rights auctions which funded metro expansion; Crossrail’s itemised contracts to minimise and expedite change orders; Milan’s long term infrastructure planning agency which works hand in glove with the MM; Copenhagen’s cheap-and-cheerful metro design; Turin’s cut-and-cover metro construction method; the MTB in Hong Kong for whom operating a rail network is a loss leader for their retail and property development business. Innovation The goal of our innovation policy is for Irish companies to be as innovative and productive as the leading European countries. Say Denmark. This takes in everything from basic research at Irish universities to the details of taxation of equity options. We believe scientific research could be much more productive. New approaches to research promise to improve matters. The core idea is to “turn the scientific method on itself”. Just as researchers experiment in the lab, metascience involves studying science itself with trials and measurements. A US National Science Foundation pilot offering top scientists greater stability and flexibility has increased productivity. In New Zealand and Switzerland, lotteries have mitigated funding bias. Meanwhile, the UK is developing metrics for research novelty. This will ensure funding is allocated to genuinely groundbreaking work. We believe Ireland’s failure to cultivate world-class domestic technology and pharmaceutical firms is a blunder. There should be a conveyor belt of talent leaving multinationals, starting businesses, selling businesses, and funding the next generation of entrepreneurs. Bored mid-career engineers and scientists at multinationals are an untapped strategic asset. We need to make it worth their while to leave their stable careers and try big things.Innovation policies we’re inspired by: the Arc Institute, built from the ground up on metascience principles, the use of ARPAs to fund specific mission-based research; the US tax treatment of equity options; US bankruptcy laws; EU Inc (the proposal not the version that has emerged from the Commission) the use of open science principles; an EU common market for digital services, Denmark’s flexicurity laws.