Exercise: housing 25% cheaper (I): build more, faster and better connected

This week's exercise comes straight from the programme: reduce housing costs by 25% without mass displacement or huge windfalls to the people who already own land. It has to cover planning, permits, density, construction productivity, infrastructure, land-value taxation, social housing, rental protection, housing benefits, empty homes, short-term rentals, transport and energy renovation, all at once. It also has to leave us suspicious of two easy answers, the NIMBY progressivism that blocks building in the name of the neighbourhood and the libertarian line that deregulation alone will do it.

We'll do it in two parts. This one covers the supply side: how Spain gets far more homes built, quickly, in the places people want to live. Tomorrow's part covers who captures the gains, who is protected while prices adjust, and what to do with the homes that already exist. The first part makes housing cheaper; the second makes sure the savings reach tenants and young buyers.

Start with where the shortage is

The Banco de España's latest annual report, as elDiario.es summarised it in June, puts Spain's housing shortfall at 750,000 homes for 2021–2025, up from about 600,000 in its 2023 estimate. That is about 3.7% of resident households, against 1.5% in Germany. In the latest year Spain formed 240,000 new households and completed 92,000 homes. Just over half the shortfall, 52.5%, is concentrated in six provinces: Madrid, Barcelona, Alicante, Valencia, Murcia and Málaga. Over the same years, home ownership among young people fell from 41% in 2007 to 22% in 2025.

That geography decides the method. A national housing problem that sits mostly in six metropolitan areas needs a plan that works at the scale of those areas. The OECD's Brick by Brick gives the reason. It finds that the more decentralised land-use decisions are, the more restrictive they tend to be, because existing homeowners press local politicians to protect the value of their homes. Coordination at metropolitan level is associated with denser cities and less sprawl, and overlapping responsibilities with more stringency and delay, since several levels of government can each veto a project.

Move one: metropolitan housing compacts

So the first move is a compact for each of the six areas, signed by the state, the region and the municipalities, with a housing target set for ten years, the land to meet it identified, and transport money attached. Our piece on a national neighbourhood plan describes how the Netherlands did this with Vinex in 1995–2005. The compact needs a trigger to bite. If a metropolitan area falls behind its target, or if its prices keep rising faster than local incomes for several years, the rules on what can be built automatically widen across the whole area. A council that wants to keep its own rules simply has to deliver its share of homes.

Move two: permits by right

The CNMC's June study on land for housing describes the chain of delays: detailed plans that can be annulled whole, piles of technical and administrative steps for urbanisation, multiple controls on each building permit, and sectoral reports from environmental, heritage and infrastructure bodies that are poorly coordinated. Its remedies are concrete. Extend sworn declarations in place of licences, scrap duplicate checks, replace negative administrative silence with positive silence for sectoral reports, require each report to justify itself, and limit the total annulment of plans.

Put together, those make building by right possible. A project that meets the published plan and the building code goes ahead on the developer's sworn declaration, and the council inspects afterwards, with heavy penalties for false declarations. Our earlier piece on three years of paperwork sets out the details.

Move three: density where the trains are

The CNMC notes that land can make up as much as 45% of the final price of a home. The cheapest way to lower the land cost of each flat is to put more flats on each plot in the places that can take them, which means close to stations and frequent bus lines. Edward Glaeser makes the case most bluntly. In Triumph of the City, as a review in Governing reports, he estimates that Manhattan could roughly halve the cost of a typical large apartment building by building as freely as Chicago. In Spain the gain lies in allowing a few more floors near transport, and in releasing the parking and plot-size rules that our piece on cost per square metre lists.

Move four: build more productively

More permits will not help much if the industry cannot build. BBVA Research found that labour productivity in Spanish construction, measured as value added per hour worked, has fallen by more than 20% since 2013, while it rose 5% across the economy as a whole. Workers under 30 made up almost 30% of construction employees in 2007 and only 10% in 2023, and vacancies for construction trades rose by 300% after 2016. The OECD notes that construction costs for new homes in the OECD-EU area rose by more than 70% between 2000 and 2019, and that Germany set up a commission on cutting building costs which produced over seventy recommendations.

A radical-centre package here mixes public and private roles. The state approves a catalogue of standard building designs that any council must accept, which lets factories build components in long runs, as our article on factory-built housing describes. It pays for dual vocational training in the building trades, tied to the compacts' volumes so that firms can count on years of demand.

Move five: bring cheaper land within reach

Alain Bertaud, whose Order without Design is on this week's list, treats a city as a labour market. In M. Nolan Gray's review in City Journal, Bertaud's point is that people will always travel well beyond their neighbourhood for work, and that the planner's job is to allocate scarce road and rail space so commutes stay short. In housing terms, every new commuter rail line that puts cheap land within a short commute of the jobs adds supply as surely as a new building. Each compact would therefore fund trunk transport, water and sewers ahead of demand on the land it designates, with the cost recovered from the value the works create, which is where part two begins.

What this half adds up to

None of the five moves alone gets Spain to a quarter off. Together they attack the parts of the price that rules and delay add: scarce serviced land, slow permits, low density near transport, an industry short of hands, and jobs out of reach. The left gets public leadership of metropolitan growth and binding targets. The market-liberal tradition gets permits by right and competition between builders. The idea that fits neither camp is the scale: housing decided where newcomers have a voice equal to incumbents, with triggers that keep everyone honest. Tomorrow we look at how to stop landowners pocketing the gains and how to protect tenants while the building catches up.