This post takes the next step. Ignore the reasons. Start from the assumption that the CSO can count, that its numbers are accurate. The highest-demand parts of Dublin really do have the highest vacancy — not despite being in demand, but whilst being in demand. What would explain that? And is the explanation strange enough to dismiss, or ordinary enough that we should have expected it all along?

Signal and noise

Every dataset contains both. Signal is the part that carries information — the pattern that means something. Noise is the random scatter on top of it: measurement error, outliers, one-off flukes, the points that don’t line up with anything. An essential skill in reading data is telling the two apart.

If you treat noise as signal — you may try and chase a pattern that isn’t there, or build a theory on a coincidence, but ultimately you’ll learn nothing useful. More importantly if you treat signal as noise — you’ll not only learn nothing useful, you’ll miss the opportunity to learn something new, dismissing something as an outlier because it doesn’t fit what you expected.

The Irish vacancy debate has managed to do both at once, and in exactly the wrong combination.

The census recorded a number — a count of empty homes — and a set of reasons those homes were given as empty. The count is the hard data: a national enumeration, trained staff, repeat visits, a consistent method. The reasons are the soft data: a category ticked for a sample, self-reported or a neighbour’s best guess, largely unverifiable. And the consensus took the hard count to be the unreliable thing — too high, must be wrong, the CSO can’t really count empties — while treating the soft reasons as the solid ground to argue from: “look, we know there were not 30k available rental properties, the data is unreliable.” The count was filed as noise. The reasons were promoted to signal. It was precisely backwards.

I will come back to the reasons later, because they reward a closer look than they got. For now, set them aside entirely and stay with the count — the part that was dismissed — because Ireland’s census happens to have unusually good data for testing whether a dismissed number is noise or signal. It reports vacancy not just nationally and by county, but in exceptionally local granular detail: by Local Electoral Area, and below that by Electoral Division. If a number is noise, it scatters as you zoom in. If it is signal, it sharpens. So let us zoom in.

The data by LEA in Dublin

Here are the top ten highest-vacancy Local Electoral Areas in Dublin, on the 2022 census-night measure — the whole-stock snapshot — with the 2021 median sale price alongside.

For context, number 10, Artane-Whitehall, matches the whole Dublin rate at 5.6%, and the lowest LEAs are out in the suburban ring — Palmerstown-Fonthill at 2.4%, Lucan at 2.8%, Firhouse-Bohernabreena at 3.2%.

The first thing that jumps out is price, and it jumps out because of one name. Pembroke LEA — Ballsbridge, Sandymount, leafy, valuable, in demand, the most expensive electoral area in the country at €800,000 — also has the highest vacancy in the county. That is not where the consensus says the empty homes should be. The dearest district in Dublin should, on the standard logic, be the one most fully occupied: every unit precious, nothing left idle. Instead it sits at the top of the vacancy table.

Is Pembroke an outlier — a single weird point, the kind you dismiss to clean up the chart, the noise, as it were? Let us zoom in further.

Pembroke, by Electoral Division

Drop below the LEA into Pembroke’s constituent Electoral Divisions, and the single high number resolves into a field of high numbers.

These are not just among the highest vacancy rates in Dublin. They are among the highest in the country — and they sit in the highest-demand square kilometres in the State. The lowest ED here, Rathmines East C at 6.5%, would still rank near the top of the all-Dublin LEA table. Pembroke is not one vacant pocket dragging up an average. It is high throughout, and it peaks at nearly 16% in the heart of Ballsbridge. Whatever this is, it is not a rounding artefact and it is not confined to a single odd corner.

By contrast, the areas with some of the lowest demand in Dublin by price carry some of the lowest vacancy. The cheaper outer LEAs — Tallaght, Clondalkin, Lucan, Palmerstown — cluster near the bottom of the vacancy table with the lowest vacancy rates in Dublin.

At first glance: high price, high vacancy; low price, low vacancy; the two move together, but inverted to what Economics 101 tells us.

But it is not that simple, and the places where it breaks are the interesting ones. Some genuinely high-demand, high-price areas sit below the Dublin average. Rathfarnham-Templeogue, comfortable middle-class suburbia at €550,000, runs at 4.4%. Glencullen-Sandyford, prosperous and well-served, is at 4.5%. If high vacancy was actually correlated to high price, these would be elevated. They are not.

When you notice the pattern is not following price cleanly — when high-priced suburbs come in low and a mixed north side inner city areas come in high — you are being told that price is a proxy for something else, and not a perfect one. The real correlation is not just price. It is spatial. It is the inner urban core versus the outer suburban ring. Inner city plus amenity rich leafy and coastal areas closest to the city have the highest vacancy.

No. 5 on the table looks at first glance as though it may buck this trend. Cabra-Glasnevin — middling on price, €395,000, on the whole not a gilded district, nor in the heart of the inner city — sits up at 7%, comfortably above the Dublin median.

Then you drop below the LEA, to its constituent Electoral Divisions, and the outlier dissolves into a gradient. The vacancy is not spread evenly across Cabra-Glasnevin. It is concentrated, and it is concentrated in one direction.

The high-vacancy EDs are the inner city, eastern end, topped by Inns Quay A and B. The lowest vacancy EDs, the Cabra West housing estates and outer Ashtown, are below the Dublin average. The split runs almost monotonically from east to west, and it is steep: the top of the LEA is at 11%, the bottom near 4%. These are a couple of kilometres apart — and one end of the LEA is empty at nearly three times the rate of the other.

(One ED earns an asterisk rather than a theory. Phoenix Park, at 9.3%, is not a residential neighbourhood in any ordinary sense — it is the Áras, the US ambassador’s residence, Farmleigh, the official state guest accommodation, various OPW buildings. Its “vacancy” is institutional, and it should be read out of the residential pattern, not into it. It is a genuine outlier, noise as it were! Set it aside and the gradient is, if anything, cleaner.)

For Cabra-Glasnevin a radial “distance from the city centre” story does not quite capture it either, because the real axis is not the GPO. It is Croke Park. The high-vacancy EDs are the streets within walking distance of the largest stadium in the country, that anyone who has ever tried to find a room for an All-Ireland final or a three-night concert run knows intimately. It is not prime residential territory. But it is prime short-let territory: central, walkable to the stadium and the city, on the rail line. The closer an ED sits to that axis, the higher its vacancy. A Dubliner does not need the regression. They need the location named, and then it is obvious.

At the scale of the Electoral Division, the census vacancy is not a puzzle at all. The vacancy is exactly where conditions combine to reward vacancy. The more hyper-local you look at these numbers, the more they make sense — which is the exact opposite of what you would expect from noise. Noise gets less coherent as you zoom in. Signal gets stronger

Economics 101

There is a phrase that gets reached for whenever the vacancy debate surfaces, to make it sound inarguable: Economics 101. High-demand areas have low vacancy. It’s almost tautological — people are more likely to live where demand is high, prices are higher there, so the incentive to leave a property empty is weaker. Prosperous cities have low vacancy; sleepy villages have empty houses.

Notice what the word “tautological” does. If the relationship is a tautology, it is so obvious it is entirely unnecessary to explain it, it cannot be wrong, and anything that contradicts it cannot be data — it can only be noise. A weird outlier. An exception that proves the rule. This approach doesn’t just predict low vacancy in high-demand areas; it pre-commits you to disbelieving any high-demand area that shows high vacancy, before you’ve even looked at it.

The belief only holds because of a buried assumption. “People are more likely to live where demand is high” quietly equates demand with long-term residential occupation. It also quietly assumes two other basic tenets of Economics 101 - (i) “ceteris paribus” or all other things being equal and (ii) human beings will act rationally in their self interest.

But what happens when all other things are not equal? A rational decision depends on real circumstances, not textbook assumptions.

If you take the position that the CSO’s count is more likely to be reliable than your own confirmation bias, the question stops being “where can I find better data?” and becomes “do conditions exist in which a rational owner would leave a high-value property empty?”

If those conditions do exist — and especially if several stack on top of one another — then a pattern of elevated vacancy in exactly the areas the census identified is not noise. It is the signal.

When vacancy is the rational choice

A dwelling can be in ferocious demand and still hold no resident on census night — if its most profitable use is not housing a household. The census counts whether a household was occupied. It does not count whether the property was in high demand. In much of the country those are the same thing. In the most valuable few square kilometres of Dublin, they come apart.

Economics 101 assumes the only way to make money from a dwelling is to put a household in it — sell it or let it to a resident — so an empty home is always money left on the table, and the higher the value, the more money, and therefore the stronger the pull toward occupation. Anything else would be irrational. That logic is sound only if letting or selling is the best available use. If a more profitable (or less risky) use of an empty property exists, the incentive inverts: high value now raises the return to not housing a household.

Changes in the market over the last ten to fifteen years have created conditions with both opportunities and risks that incentivise vacancy.

A rising market rewards waiting. When prices are climbing, an empty property held through the rise captures the capital gain without the frictions, obligations, or sitting-tenant risk that come with letting it. Vacancy is not a cost to be minimised; it is a position being held.

RPZ rules made vacancy a way to reset the rent. Through the period the census measured, under the Rent Pressure Zone regime, permitted rent increases were capped and a below-market tenancy could not simply be repriced. The cap attached to the tenancy, not the dwelling. For an owner sitting on a long-standing under-market let, the route back to market rent ran through a period of vacancy. The regulation designed to protect tenants made an empty property, for some owners, the rational waypoint to a higher return. The incentive to hold a unit empty was written into the legislation.

This incentive is strong whether you intend to let it in the future or sell it on the open market. In all the areas of Dublin that have high vacancy rates, properties that have no RPZ restrictions will carry a premium.

And meanwhile, there is an alternative income that beats the rent. This is the one that has been measured, and the figures are not subtle. The ESRI’s 2025 profiling of short-term lets across Ireland — Profiling Short-term Let Usage Across Ireland, Devane, Kelly and Slaymaker — set the daily short-let price against the daily private-rental price and estimated how many nights a property would need to be let to match a full month’s rent. In Dublin City, the answer is eight to ten nights a month. Let a property for roughly a third of the month on the short-term market and you match what a full-month tenancy would pay. The report is careful that these may not be like-for-like properties, but it draws the obvious conclusion itself: the figures highlight the attractiveness of the short-let option. An owner clearing the equivalent of a month’s rent in eight nights with no legislative obligations has little incentive to take a long-term tenant — and every reason to leave the property empty, between guests.

It is unlikely to be a coincidence that the highest vacancy rates in Dublin are in exactly the same areas in which you’d expect to find the highest demand for short term lets eg North and South Inner City, Ballsbridge and Croke Park. Rathfarnham and Stepaside have comparatively low vacancy rates. They are expensive areas, hugely popular with affluent middle-class families - less so with airbnbers.

None of these conditions is contested. The rising market is observable. The RPZ rules are on the statute book. The short-let economics are an ESRI table. The only question is what they add up to.

The vacancy data shows what some of the most-wanted stock in those neighbourhoods is actually doing — which is sitting empty on census night, because in the highest-value, highest-amenity locations, the alternatives to long-term occupation are most profitable precisely where demand is most intense. Demand does not guarantee occupation. In specific locations, it competes with it.

Back to the reasons

Now the reasons can be picked back up, because zooming in has clarified what they mean.

The census count was dismissed in large part because the reasons didn’t smell right — among them, 33,653 properties recorded as vacant while enumerated as “for rent.” That looked absurd. There weren’t thirty thousand empty rentals sitting on the open market. That is undisputed. So the count must be inflated. The reasons discredited the number.

But turn it around. The count is the signal; the reasons are the noise. And once you know the conditions are present, the reasons start to look less like proof the count is wrong and more like proof the reason field is. Consider what each condition looks like through the census form. A short-let is, by definition, enumerated as a rental — it is a property let to people, just not for long. A unit deliberately held idle while the rent resets before a new tenancy is, by definition, enumerated as a rental. A property kept empty until it can be sold free of RPZ legislation is, by definition, enumerated as “for sale.” A property whose owner intends to sell but feels it is rational in a rising market to wait until next year, is by definition, enumerated as “for sale.” Every one of those is a rational reason to be empty on census night, and every one of them lands in a reason category — “for rent,” “for sale” — that the discourse then pointed to as evidence the home wasn’t really vacant.

By getting caught up in the reasons, we dismissed the signal as noise, and promoted the noise to signal. The count was telling us how many homes were empty. It was right. The reasons were never built to tell us why with any reliability, and they didn’t — but they were the part everyone reached for.

Signal or noise

So here is the choice, laid out plainly, because the whole thing comes down to it.

On one side: the high-vacancy areas are outliers. Weird districts. Exceptions to a rule that is basically tautological. The data has a general shape nationally — high demand, low vacancy — Co Dublin compared to Co Leitrim confirms this. Anything that contradicts this are the scattered points that don’t fit, explained away one at a time and not worth a theory.

On the other side: the conditions that would make vacancy rational in high-value central Dublin areas are all present, documented, and uncontested. A rising market that rewards holding. A rent-cap regime that made vacancy the route back to market rent. A short-let market that matches a month’s rent in eight to ten nights, concentrated in the central LEAs, skewed toward the scarcest small units. Stack those up, and a pattern of elevated vacancy in exactly those areas is not the exception to the rule. It is exactly what you would expect.

Ask which is more likely. That several of the most reliably measured, most desirable, highest-value districts in the country happen, coincidentally and individually, to be statistical flukes — flukes that line up along a stadium and a coastline and a rail line, and that get more coherent the closer you look. And that this coincidence scanned over the period of two censuses.

Or that they are showing you, accurately, rational human beings making rational decisions based on local market circumstances.

The consensus says the first. It says the count must be misleading, because Economics 101, and the areas that contradict must be noise. But all other things are not equal. Look at the conditions, look at the data, zoom in as the signal gets stronger and stronger. The second answer is not just available. It is the more economical and plausible one.

Two consecutive censuses told us something. The last post argued we should believe the count. This one argues that once you do, the explanation is not exotic.

Please stop the obsession with electricity meters, and let the Central Statistics Office get on with what it does best - providing us with reliable and accurate data.

It’s not their fault if those who interpret the data struggle to separate the noise from the signal.

This post is about the opposite reflex. It is about what happens when a figure points the wrong way — when it contradicts what we already believe. Then the machinery runs in reverse. The number is questioned, picked at, denigrated, dismissed and ultimately set aside, and again not through any rigorous analysis but through the steady repetition of congenial doubt.

The two failures are the same failure wearing opposite clothes. Groupthink does not care about the quality of the evidence. It simply asks: does this number agree with what we already believe? If it does, it becomes hard fact, however soft its construction. If it doesn’t, it becomes unreliable, however robust its construction.

The credence attached to a figure is set by its conclusion, not its method. And the standards are not merely inconsistent — they are inverted. In the examples here, the more assumption-laden the figure, the more firmly it is believed; the more rigorously it was measured, the more readily it is discarded. Ireland has decided to trust its guesses and to doubt its measurements.

There is no cleaner example of this than the dismissal of consecutive census vacancy figures. For almost a decade the ability of the Central Statistics Office to accurately count the number of vacant properties has been repeatedly questioned over two censuses.

A number that pointed the wrong way

Prior to 2017, no serious person ever questioned whether the Central Statistics Office could count. Counting is what it does, it’s raison d’être. Its output carries an implicit authority, and ministers and media would reach for that authority whenever it suited them.

Then, in 2016, the CSO counted something inconvenient. The census recorded over 200,000 vacant homes and a national vacancy rate of 12.3 per cent — stripped of holiday homes, they found 183,312 vacant, habitable dwellings, a rate just over 9%. The number was awkward. The figure did not automatically disprove the housing emergency. But it complicated the simplest version of it. It implied a stock of empty homes that sat uneasily beside the political story of pure scarcity. And so, for the first time, the CSO’s ability to count was put on trial — not strictly its methodology in any technical sense, but its answer.

The reaction was not to ask what the figure meant. It was to find a better one. Within a few months of the CSO releasing the figures, the discourse had a more congenial measure to hand - GeoDirectory - and, within a few years, another - the CSO Electricity meter count. The Census vacancy figure — the product of the largest and most systematic enumeration exercise the State performs — was effectively retired from the conversation, replaced by numbers that told people what they wanted to hear.

How the yardstick was switched

You can watch the switch happen in real time, in the same publication, by the same authors, across three editions.

GeoDirectory — the An Post and Ordnance Survey address database — publishes a residential buildings report twice a year. In its Q2 2016 and Q4 2016 editions, the vacancy section simply reports the Census figures as authoritative. There is no rival measure. The CSO count is the count.

Then the Census figure came under public attack upon release in April 2017. And in the next edition of the report, August 2017, GeoDirectory introduced its own vacancy measure — one that came in at roughly half the Census figure. It is worth quoting the explanation from the report in full, because of what it does (emphasis added):

“Based on the above definitions, GeoDirectory report a vacant stock of 96,243 address points or units in June 2017, while the 2016 Census reported a vacant stock of 183,312 address points or units, as of April 2016. Thus the GeoDirectory figure is around half the Census figure, which is a substantial difference, of the order of 87,000 dwellings. The average vacancy rate across the State is 4.9 per cent, according to GeoDirectory (Figure 10), compared with 12.3 per cent, according to the Census of Population.

Drilling down further, however, it is possible to explain some of this substantial difference. The CSO has provided some data on the reasons why vacant dwellings were vacant at the time of the Census of Population for a small sample of vacant buildings (i.e. around 57,000 dwellings or close to one-third of the total). For this one-third of vacant dwellings, they include dwellings classified as for sale (10,948 dwellings), for rent (10,350), owner in nursing home (4,165), renovation work underway (3,678), owner in hospital (1,469), and owner with relatives (847).

Some of these categories could be construed as dwellings which might not normally be classified as vacant in the context of vacant long term, but would represent more of a transition or temporary vacancy rate, i.e. while properties are waiting to be sold or rented out. In the aggregate they represent a total of around 31,500 properties out of the 57,000, or 55 per cent, implying that 25,500 of this total would be deemed to be vacant. As these explanations were only provided for one-third of vacant dwellings, (if it is assumed that 55 per cent of the remaining two-thirds were similarly classified, leaving 45 per cent as representing the true vacant total) this would reduce the CSO figure for the number of vacant dwellings considerably to around 83,000, which would be closer to the GeoDirectory figure of 96,243.”

Read what is happening there. The Census count is not being corrected. It is being explained downward until it lands near GeoDirectory’s own figure. The method is to take the reasons the CSO recorded for one-third of vacant homes, decide that more than half of them — the ones for sale, for rent, mid-renovation, owner in a nursing home — are not “really” vacant, and then assume the same proportion holds for the two-thirds nobody has data on. An assumption applied to a sample, extrapolated across the unmeasured remainder, to arrive at a number that chimes with the expectation. This is the deficit’s reverse-engineering, run on vacancy.

And notice precisely what is being struck out as “not really vacant”: homes for sale, homes for rent, homes between uses. That is not noise. That is the single most important category of vacancy there is. A functioning housing market requires a layer of empty homes in transition — properties available for someone to buy, to rent, to move into — because that is the slack that lets people move at all, the liquidity that makes the market function. Strip it out and you have not corrected the count. You have deleted the most important thing a point in time vacancy figure exists to capture, and called the deletion an improvement.

Of course the for sale and rent reasons recorded for vacancy in the census do not correlate to what was actively listed for sale and rent on the market at the time. That alarm bell should have been the very reason to drill down into the data further and seek out what it was telling us. Instead, in as far as the discrepancy was even acknowledged, it was used as evidence to cast doubt on the accuracy of the count. We never questioned the accuracy of the reasons, and thus missed the opportunity to learn something from the answers.

GeoDirectory’s measure, incidentally, rests substantially on whether a dwelling is receiving post. The Census used trained enumerators making repeat visits, looking for furniture, cars, accumulated mail, an overgrown garden, and asking the neighbours — and was instructed that one or two visits were not enough to call a property vacant. The report sets both methods out on the same page. It then treats the postman’s round as the corrective to the enumerator’s count. Once GeoDirectory’s lower figure had become an acceptable yardstick, the caveat against the Census quietly disappeared from later editions. The job was done.

Census 2022 suffered the ignominy of having to include its own caveat, which in itself became a citable reason to cast doubt on the accuracy and usefulness of the figures. For the record Census 2022 found a national vacancy rate of almost 8%.

The CSO also added its own electricity-based measure in 2022 — a dwelling drawing almost no power across four consecutive quarters. It is a good answer for the right question, and an honest one. But it measures one specific thing: long-term, sustained emptiness. Like GeoDirectory’s, it is designed to strip out exactly the transitional, frictional vacancy that the Census captures. None of this would matter if everyone kept the measures straight. The problem is what came next: the long-term measures became the headline “vacancy rate,” and were then used to answer questions they were never built to answer.

Apples and oranges

Consider a piece earlier this year from the think tank Progress Ireland, arguing that land hoarding and speculation are not major drivers of the housing crisis. To make the case, it turns to vacancy, and again it is worth quoting at length:

Nevertheless, we do have a lot of data on the vacancy rates of dwellings. Data about vacancy rates (or “long term” vacancy, ie four consecutive quarters) comes from the CSO. If a dwelling is using less than 180 kilowatt-hours per quarter, then it is considered vacant.

If the modern Georgist idea is right, that hoarding valuable sites is a widespread and consequential problem, then we would expect to see high levels of vacancy. More than that, we would expect to see high levels of vacancy in the highest demand areas. Withholding an unpromising piece of land would make for dubious speculation.

The second thing to say about vacancy is some level of vacancy is not just expected, but is desirable.

Policymakers sometimes refer to a “healthy” vacancy rate of about 5 per cent. The reason why a functional system should have some vacancy is that people move around. If all homes are filled, then mobility falls. That means, people can’t move to better jobs, to be nearer family, or to downsize. A functional system will have some mobility as people move jobs or relocate for other reasons.

The third thing to expect is that areas with high demand will have lower levels of vacancy than areas with lower demand. A healthy system will match people to empty homes faster in areas with higher demand. You should expect prosperous cities to have lower vacancy rates than sleepy villages.

As George said, unusually high levels of vacancy is a symptom of “hoarding” or speculation. So, how is Ireland faring?

Across Ireland, vacancy levels have been falling.

Dublin in particular has unusually low levels of vacancy. In South Dublin, the vacancy rate hovers around just 1 per cent. In Dún Laoghaire, vacancy levels are at just 0.9 per cent. As you would expect, vacancy levels outside of Dublin and Cork city are typically higher.

It begins admirably precise about which measure it is using:

“Data about vacancy rates (or ‘long term’ vacancy, ie four consecutive quarters) comes from the CSO.”

Long-term vacancy. Stated plainly, defined exactly. So far, this is the disciplined thing to do. But watch what the same passage builds on top of that definition:

“Policymakers sometimes refer to a ‘healthy’ vacancy rate of about 5 per cent. The reason why a functional system should have some vacancy is that people move around. If all homes are filled, then mobility falls... A functional system will have some mobility as people move jobs or relocate for other reasons.”

This is the switch, and it is so smooth it is almost invisible. The “healthy 5 per cent” being invoked is a benchmark for total vacancy — it describes precisely the moving-around, the transitional slack, the homes between tenancies and sales. It is a frictional concept. But the figure being measured against it is the long-term rate, the four-quarters-dark rate, which is built to exclude exactly that mobility. The argument compares a number that captures only the dead stock against a benchmark that describes the living, breathing churn — and concludes the market is short of slack. The two are not the same quantity. You cannot measure the breathing with an instrument designed to filter the breathing out.

The same move runs through the rest. The piece says, correctly, that “areas with high demand will have lower levels of vacancy than areas with lower demand.” On the long-term measure, that holds — South Dublin shows around 1 per cent. But include the frictional vacancy that the Census counts and the highest-demand parts of Dublin do not uniformly show the lowest vacancy, quite the opposite. The highest-demand areas of Dublin have the very highest vacancy rates, some in excess of 10%. The conclusion depends entirely on keeping the frictional layer out of frame.

And then the international comparison, which is the tell:

“The levels of vacancy in Dublin are unusually low by international standards. New York, Paris, and London all have higher vacancy rates than Dublin.”

The figures for Paris and New York are total vacancy figures — INSEE and the US Census Bureau count all vacant dwellings, frictional and long-term together. The note on the chart acknowledges that rates are measured differently, but this is not even close to comparing like with like. It is comparing Dublin’s dead-stock-only number against other cities’ all-vacancy numbers and declaring Dublin unusually low. Compare apples with apples — total against total — and the gap narrows substantially or vanishes. The “unusually low by international standards” claim survives only on the strength of measuring Dublin one way and Paris and New York another.

This is not a swipe at Progress Ireland, nor the author of the piece. It’s just one example of a pervasive error. For other examples see the Housing Commission, the Dept of Finance, respected housing economists, Irish Times opinion pieces and letters to the editor. It’s everywhere.

The recency of the Progress Ireland example shows the culmination of a decade of misguided discussion about something which originally was little more than political inconvenience and public incredulity. Now we have a housing expert at an influential think tank that has the ear of the government unwittingly comparing apples with oranges, and using that comparison to state with authority “the levels of vacancy in Dublin are unusually low by international standards.”

Why a good number feels wrong

There is a reason that able, honest people make this error, and the geographer Rob Kitchin, in a paper titled “Data debates in urban development: the data politics of facts and counter-facts”, calls it out precisely:

“Property and planning analysts and market actors usually possess a wealth of phronesis and metis that, when confronted with data that presents a different picture to their grounded experience, can spark feelings of unease or bewilderment and seed a sense of distrust in a dataset.”

Phronesis and metis are the practical, experiential knowledge that comes from working in a field — the estate agent who knows there is nothing to rent, the official who knows the emergency is real. That knowledge is valuable. But Kitchin’s point is that when a measurement contradicts it, the instinct is not to ask whether the experience is incomplete. The instinct is to distrust the measurement. The number feels wrong, so it must be wrong.

That is the whole story of the Census vacancy figure. It conflicted with what everyone knew — that there was nothing to rent or buy, that the crisis was acute — and so it was assumed to be flawed. Nobody stopped to consider that both things could be true at once: that the stock on the market could be desperately tight and that the Census was accurately counting a large number of homes empty on a given night, because those are answers to two different questions. The unease did its work. The figure was distrusted into irrelevance, and softer numbers that matched the grounded experience took its place.

The acid test

Here is the test. To dismiss the Census vacancy count, you have to commit to a single proposition: that the Central Statistics Office — a body whose only ideology is accuracy — cannot accurately count empty houses. Not “cannot interpret why they are empty.” Cannot count them.

Say it plainly and it falls apart. Nobody will seriously defend the sentence ‘the CSO cannot count’ once it is written down. And yet it is the load-bearing assumption beneath a decade of housing commentary. We have preferred, in effect, to trust the postman over the census-taker — not because the postman’s method is better, but because his number was more comfortable.

The issue is not whether the CSO can count. It is whether commentators are entitled to treat a narrower measure of long-term vacancy as if it disproves a broader Census measure of total vacancy.

None of this is an argument that there is a silver bullet waiting to be fired. It is an argument about making sure we understand the root causes of the problem, eliminate blind spots, and resist groupthink. It is an argument about which evidence we are allowed to take seriously, and on what grounds.

The Census is a measurement. It deserves to be engaged with as one — interrogated, yes, but not waved away because its answer is unwelcome. The Census figure is not self-interpreting. It does not tell us why homes are empty, how long they have been empty, or how many could be brought back into use. But it does tell us that they were counted as empty under a consistent national method.

The deficit and the cost figures are estimates, and deserve to be weighted as estimates, not laundered into facts. We have it exactly backwards.

And once you see that, a larger thing comes into view. Almost every confident proclamation about how to solve the housing crisis rests on assumptions — about population growth, migration, household formation, how many people want to live alone or with others. These are not measurements. They are projections, and a projection is only as good as its assumptions, which in this debate were chosen to point in a congenial direction while the measurements that might have disciplined them were set aside. The solutions everyone argues about are downstream of those assumptions. Which means the debate that looks like a debate about solutions is really an unexamined debate about assumptions — and that debate has scarcely happened.

Start instead from a different assumption — that the CSO can count, that the measurements are real — and a different diagnosis comes into view, and with it a different set of questions. Not because that diagnosis is certainly right; it rests on its own assumptions, and they would need their own scrutiny. But because it reveals that the confident answers everyone repeats were never forced by the evidence. They were one branch of a tree, and nobody walked down the others, because the measurement that pointed toward them had already been dismissed on the grounds that the national statistics office cannot count.

Two consecutive censuses were telling us something we did not understand, and we have ignored that for a decade while the problems have grown progressively worse.

The first step to solving those problems is to stop pretending we already know all of the answers.

In this instance it is the affordability gap based on the ‘real’ cost of construction for a three-bed semi-detached house reported in the Society of Chartered Surveyors Ireland’s Real Cost of New Housing Delivery 2023

The SCSI produced these costings based on “some of the most up-to-date construction and market data” and the report says its data “ensures strong, evidence-based policy recommendations for housing delivery across the country”.

On average, the cost of delivering a three-bedroom semi-detached home ranges from €354,000 in the Northwest region, to over €461,000 in the GDA. The overall average delivery cost of a new three-bedroom semi-detached house is 30% higher in the GDA when compared to the lowest delivery cost of the Northwest region

So far, so good. Exactly what one would instinctively expect. Land costs would be significantly higher in the Greater Dublin Area (GDA, which combines Dublin, Kildare, Meath and Wicklow). Labour would also understandably be higher, as would a myriad of other costs.

The regional headline figures of build costs looked about right and in the rush to justify or judge government policy everybody ran with them. Straight into press releases, opposition speeches, industry submissions and explainer pieces, repeated as hard evidence of the cost of building a home in Ireland.

The 2023 report attempts to base costings on a standardised house — a 114 square metre, three-bedroom semi-detached, built to a standardised specification — and prices it across seven regions, from the Greater Dublin Area down to the Northwest. It is a genuinely useful exercise, specifically intended to enable regional comparison and a more ambitious one than the SCSI’s earlier (Dublin-only) reports. The output everyone quoted is the headline total: a 3-bed semi costs €461,437 to deliver in the GDA, falling to €353,896 in the Northwest — a spread of about 30% top to bottom, or about 15% on the “hard costs” of actually building the house.

That spread is the expected story. If the variation lived only at the level of the totals, there would be nothing questionable at all.

But the report doesn’t stop at totals. It publishes the full build-up for each region — every line, to the euro. And once you line those build-ups up side by side, the tidy 15% spread on the subtotal dissolves into something much stranger underneath.

And nobody appears to have looked closer at the component costs that make up the total. These costs should have at the very least given pause for thought.

There are some odd regional differences in the costs of fittings, finishes and services. They are standardised precisely to enable comparison, but direct comparison raises more questions than answers. “Fittings” comprise of a fitted kitchen with a formica worktop, no appliances, and a wardrobe to the main bedroom. According to the report, that standardised specification costs €6,498 in the Northeast region and €19,950 in the Northwest. Same kitchen. Same wardrobe. Three times the price.

The numbers that don’t make sense

Here are the hard-cost components — the physical building of the house — for the same specified 3-bed semi, region by region.

Table 1: Hard-cost components of an identical 3-bed semi (€), by region

Look first at the bottom row. The subtotal varies by about 17% from cheapest to dearest — the credible, expected story. Now look at Finishes, Fittings and Services, and the picture inverts.

• Fittings range from €6,498 (Northeast) to €19,950 (Northwest) — a factor of three.

• Finishes range from €26,912 (GDA) to €50,472 (Northeast) — nearly double, and pointing the “wrong” way, with the supposedly cheaper Northeast the most expensive of all, Dublin the cheapest.

• Services range from €20,338 (GDA) to €34,697 (Galway) — a swing of 70%. And again Dublin is the cheapest. Totally counterintuitive.

These are not abstract cost categories. The report tells you exactly what is in them. Fittings is “fitted kitchen with formica worktop, no appliances, and wardrobe to main bedroom”. Finishes is “triple-glazed windows; paint grade softwood doors, skirting and window boards internally; paint-finished walls; and, tiling to bath and shower rooms.” Services is “quality-grade sanitary fittings; solar panel; heat recovery; heat pumps; underfloor heating/radiators; plastic electrical fittings and pendants throughout; and, wired for alarm.”

These are deliberately standardised items, to a standardised specification, in every region. A wardrobe is a wardrobe. Tiling is tiling. The materials are bought on national, often international, markets. If anything, labour to finish and fit these items in the Northwest is cheaper than in the Greater Dublin Area, not three times dearer.

So why does the same fitted kitchen and wardrobe cost €6,498 in Louth and €19,950 in Mayo? There is no physical reason it should. And here is the quietly important part: when you sum the lines, these wild swings largely cancel out, which is why the subtotal still looks sensible. What one region’s surveyors appear to have booked under “fittings,” another’s seem to have folded into “finishes” or “completion.” The total survives; the components do not.

That points, fairly strongly, to the variance in part being an artefact of how the data was assembled — a survey of around 80 separate developer submissions, priced and categorised by different chartered quantity surveyors who allocate costs to headings inconsistently — rather than a real signal about regional building costs. But I want to stress: I don’t actually know that. Nobody does. The report doesn’t explain the divergence, and as far as I can find, in almost three years nobody ever asked it to. Which is the whole point of this piece.

Why it matters: the viability table

If this were just a curiosity in an appendix, it might seem pedantic to question it. It isn’t, because the report uses these figures to do something consequential: it calculates, region by region, whether building a house is “viable” — whether the delivery cost is covered by the market value. And those findings on viability have been used to underpin, justify and extend government policy and spending, specifically Help to Buy, First Home Scheme and development levy waivers.

On the report’s total figures, the headline that travels from this is seductive and tidy: new homes only stack up in the most expensive markets; everywhere else, building loses money. That is a powerful claim, and it feeds directly into the argument that the State must step in to “bridge the viability gap.”

But that €2,599 surplus in the GDA is smaller than the unexplained variance on a single hard-cost line. The fittings figure alone moves by nearly €13,500 between regions for no obvious reason. A viability result that turns on €2,599 is being computed to a precision the underlying data cannot possibly support.

If these figures were produced to make a case for private market financing or investment rather than government policy, any entry level accountant would take one look at them and say “Your viability problem here is obvious. Stop sourcing your tiles in Louth, buy them in Dublin instead.”

So what does that look like? I took the three components that one would expect not to vary significantly by region — Finishes, Fittings and Services — and standardised them: I assigned every region the lowest value any region recorded for each (the cleanest, if deliberately aggressive, way to strip out the inconsistency). Then I let that flow through the rest of the build-up. This matters, because the report calculates developer margin as a flat 15% of costs and VAT on top of that — so when you reduce the cost base, the margin and the VAT charged on it fall too. I used each region’s own implied rates, so I’m not importing any assumption of my own.

Here is what happens.

Standardising three line items that should not vary by region flips the picture. The story becomes “GDA, Galway and the Northeast are viable, and the Northwest is essentially at breakeven.” Every remaining deficit roughly halves. And these aren’t trivial adjustments: the Northeast moves by over €47,000, because its €50,472 finishes figure — the single largest outlier in the dataset — was carrying both a 15% developer margin and VAT on top.

There is a necessary caveat here. Taking the lowest value is the most aggressive defensible choice; using the median would move everything in the same direction but less far. The point of the exercise isn’t to claim these are the true costs — I am not a QS, and cannot be sure of the true costs any more than the next layman. The point is that the viability conclusion is extraordinarily sensitive to component figures the report itself never justifies. Change numbers that have no physical reason to differ, and the headline finding reorganises itself.

The actual problem

None of this is an accusation against the SCSI of acting in bad faith. Lobbyists will lobby. They published their methodology, they were transparent that the figures came from member submissions, and they explicitly caution that the viability table “should not be interpreted as regions that are financially unviable for new homes development.” They did, in fairness, flag their own caveats.

The problem is what everybody else did with it. The €461,437 figure, and the “only Dublin is viable” framing, were picked up and recirculated as established fact — by industry bodies welcoming it as objective evidence, ministers justifying billions in subsidies, opposition politicians wielding it against the Government, government parties wielding it against the opposition, opposition manifestos using it as basis for their alternative affordable housing costings, economists in media op-eds and so on and so on. It became a fixed point that all parties reasoned from. And not one of them, as far as I can establish in almost three years of looking, scrutinised it closely enough to notice and question why the same wardrobe should cost three times more in one county than another.

This is the bit I keep coming back to. The information needed to question the figure was never hidden. It was printed in the report in plain numbers. The €19,950 fitting cost and the €6,498 fitting cost are on facing pages. Anyone who read past the headline to the second table would have seen it. The failure here isn’t in the production of the number — it’s in the reception of it. A figure becomes “true” in Irish housing not by being checked, but by being repeated.

And because it was never questioned, we are now stuck. I can show you that the components don’t make physical sense. I can show you that standardising them moves the conclusion. What I cannot tell you — what nobody can tell you — is why the figures look the way they do, whether the variance is pure classification noise or whether some of it is real, and therefore how much weight the viability conclusion can actually bear. That question can only be answered by the people who compiled the data, and it will never be answered if nobody is sufficiently curious to ask.

This blanket acceptance of figures that don’t add up is endemic in all aspects of commentary, debate, analysis and policy on the housing crisis.

We see a figure that confirms what we already think, and accept it blindly. We don’t ask the right questions because we think we already know the answers. A figure becomes ‘true’ in Irish housing not by being checked, but by being repeated.

NOTE: There is a second oddity in the same tables, and it is purely a matter of how the report is presented. In Section 2, the hard cost figures for two regions — the Greater Dublin Area and the Northwest — are printed in ordinary roman type. The figures for the other five regions are italicised. There is no key, footnote or note anywhere in the report explaining the distinction. In cost reporting, italics are conventionally used to flag figures that are estimated, derived or carried over rather than directly measured — it is unclear whether that is what is meant here, because the report doesn't say, and the pattern doesn't resolve neatly either way (the Northwest is roman, yet it is the region carrying that €19,950 fittings figure). The point is narrower and, by now, familiar: it is just another example of something that should stop a careful reader and prompt a question.

This week’s post is about problems with the methodology used in arriving at the Housing Commission deficit — the estimate of an approximately 212,500 to 256,000 housing stock shortfall that influences the entire Irish housing demand discussion framework.

There appear to be three methodological problems that compound to inflate the figure.

Problem one: the benchmark has no Irish precedent

The Housing Commission methodology that produces the suppressed-household-formation figure compares Ireland’s current young-adult headship rate against benchmarks drawn from other countries and from surveys of ideal housing preferences: respondents to the Commission’s RedC survey who said they would prefer different living arrangements if cost and availability were not limiting factors. The gap between current Irish arrangements and these comparators is treated as the measure of suppression.

Ireland’s actual headship rates under historical conditions of adequate housing supply are measurable and on the public record. In 1991, during a period of net emigration when housing was not the binding constraint, the headship rate for 20-24 year olds was about 14 per cent. By 2011, after the crash, when Ireland had a demonstrable surplus of housing stock — when ghost estates had become a national symbol and prices had collapsed by half from peak — the rate was 18.5 per cent. In 2006, at the height of the Celtic Tiger when around 90,000 dwellings were being completed annually, it was around 19 per cent. The Irish rate for this cohort, across very different conditions but consistently under adequate or excessive supply, has sat in a narrow band between 14 and 19 per cent.

The Commission’s methodology effectively benchmarks against rates Ireland has never empirically produced. There is no period in modern Irish history in which the country has produced anything like the headship rates that the underlying-supply benchmark implies, regardless of housing supply conditions. The methodology is therefore not measuring suppression of household formation against an Irish baseline; it is measuring the distance between current Irish arrangements and an aspirational benchmark imported from elsewhere.

The aspirational figure found closer to home in the survey of young Irish adults is also problematic.

For survey respondents aged 18-39, their current household circumstances were not consistent with their preferences. Arrangements are currently over-represented, particularly living with parents (26% current vs 4% preferred). At the same time, other living arrangements are under-represented in current living arrangements, such as living with one’s partner (20% current vs 39% preferred). Two-in-five 18-39 year-olds would like to live alone or with a partner. However, just 26% of respondents in this age range currently do. Conversely, while 48% live with parents or with friends/others, just 15% would choose to do so.

The methodology is worth examining in plain language, with a simple albeit somewhat facetious analogy.

Imagine the Society of the Irish Motor Industry conducted a survey asking respondents what car they would ideally choose to drive if cost and availability were not limiting factors.

The survey would inevitably identify a large gap between current vehicle ownership and stated preferences. SIMI could then publish a report calculating Ireland’s automotive deficit — the number of additional Ferraris, Bentleys and high-end SUVs required to close the gap between what people currently drive and what they say they would prefer to drive. The report would be methodologically internally consistent with the Housing Commission report. It would also be useless as a guide to forecasting demand for cars.

The Housing Commission’s deficit calculation rests on a structurally similiar idea. The empirically grounded Irish benchmark — what young Irish adults actually choose when housing supply is adequate — sits substantially closer to current arrangements than the discourse acknowledges.

In fairness to the Housing Commission, they are transparent about this methodology and the fact that the estimated deficit is an aspirational figure - what would the ideal look like?

The problem is that nobody else is acknowledging this idealism, and the deficit figure is routinely reported, quoted and referenced as a realistic measurement of the shortfall in the housing stock that is causing the current problems. Thus it has become the basis for ever increasing new build housing targets.

Problem two: the contemporary figure does not mean what it appears to mean

There is a second layer to the over-statement, and it is documented not by outside critics but by the ESRI. The 2024 ESRI paper Household size in Ireland: Stylised facts and cross country trends by Conor O’Toole and Rachel Slaymaker — both of whom advised the Housing Commission — examined the Irish EU-SILC data that produces the headline figure of young adults living with parents and flagged a specific methodological problem.

The Irish EU-SILC data series changed its definition of a household in 2020. The pre-2020 definition was address-based: a household consisted of the people living at a given address. The post-2020 definition is based on shared income and expenditure: people who pool resources are treated as a single household regardless of where they physically live. Under the new definition, Irish students living away from the family home but who are financially supported by their parents are coded as members of the parental household. Under the German methodology, the same young person is coded as an independent household.

The ESRI authors are explicit about the implication. They note that the Irish EU-SILC figure jumped sharply in 2020 and 2021 — from 6-8 per cent to 10.5 per cent for the share of households containing an adult child aged 25-34 — and that the increase “may at least partly result from this change in the household definition.” They also note that Census 2022 figures, which use a different methodology, are meaningfully lower than the EU-SILC figures imply.

The comparison that anchors the suppressed-formation argument — the gap between Irish and German young-adult independence rates that has appeared in every serious piece of housing commentary in Ireland for two years — is therefore not measuring what it appears to measure. An unknown but potentially meaningful share of the Irish figure is an artefact of how Irish students with parental support are now coded, and would not appear in the equivalent German figure. The comparison that has driven the deficit discourse is comparing figures constructed under different definitional conventions.

Problem three: the high Irish household size is mostly about children, not adults

The observed high average household size in Ireland compared to other countries is discussed in the Housing Commission report, and ultimately the report’s conclusion is:

“key drivers of household size such as fertility and longevity do not indicate any difference in trends that would explain Ireland’s divergence.”

This is directly contradicted by the Slaymaker/O’Toole paper, working through the demographic components of Ireland’s high average household size. Their research produced a finding that is genuinely surprising once stated plainly. Ireland’s outlier status on household size is driven primarily by the high number of children per household, not by young adults.

“We find that Ireland has a high average household size on a cross-country basis. However, this appears to be strongly influenced by demographics, with high fertility rate, younger population and thus high share of households with children important factors in explaining the cross country trends”

The paper explicitly disaggregates household size into adults and children, and the result is that when you look at adults per household, Ireland is no longer an outlier. It sits just above the middle of the European distribution. Ireland’s distinctiveness shows up entirely in the children-per-household figure, where it is the highest in their 16-country sample throughout the period studied.

The paper also runs cross-country regressions controlling for demographic and economic factors. When you control for the share of households with children, marital status, age of household head, and economic variables, the difference between Ireland and the middle group of European countries — Austria, Belgium, Germany, Finland, France, Italy, Netherlands, Sweden, UK — essentially disappears. Ireland’s outlier status is explained, in the main, by demographic structure rather than by housing scarcity.

This is a finding that should have, at the very least, formed part of the discussion. Instead it has been totally ignored.

The Housing Commission’s deficit calculation rests on the premise that, in April 2022, Ireland’s household size of 2.74 was artificially elevated by housing scarcity, and that under adequate supply conditions it would have fallen to 2.4.

The ESRI’s analysis disputes this, finding that the headline household size figure is mostly a function of Ireland having higher fertility and therefore more children per household than its peers. This is not housing scarcity. It is demography. A country with a younger population structure and higher birth rate will have higher average household size for reasons that have nothing to do with whether young adults can find apartments to rent.

The implication of the ESRI findings is that the Commission’s underlying-versus-observed household size distinction is largely measuring the wrong thing. It looks at Ireland’s elevated household size, attributes it to suppressed household formation, and calculates the housing deficit required to close the gap. The ESRI’s analysis suggests that the great majority of the gap is not suppressed household formation at all. It is Ireland having more children. The suppressed-formation component is real but much smaller than the headline figure implies.

It is necessary to be explicit on the timing of these publications. The ESRI paper was published shortly after the Housing Commission’s report, and perhaps the Commission did not have access to this specific methodological caution when it finalised its analysis. I’m not suggesting that the Commission deliberately ignored available conflicting research.

The problem is that mainstream Irish housing research published less than a month after the Housing Commission report, directly contradicts the findings of the Housing Commission, and nobody has even asked the simple question:

Who is correct? The Housing Commission or the ESRI?

The implications of the answer are hugely significant. If the ESRI are correct we are fighting the wrong battle.

For two years, every serious projection of Ireland’s housing requirement has rested on the same analytical move. The Housing Commission, in its 2024 report, established it. The Central Bank adopted it. The ESRI works with it. The Department of Finance’s recent Future Forty report extends it to 2040. Government targets are calibrated to it. The recent national extension of Rent Pressure Zones cites it. It is now the load-bearing assumption underneath the largest sustained capital allocation in the Irish economy.

The move is the concept of pent-up demand: the idea that Ireland has a deficit of housing represented by moves and transactions that would have taken place in conditions of adequate housing supply but did not, because supply was insufficient. The Commission estimated this stock deficit at between 212,500 and 256,000 dwellings as of April 2022. The figure has propagated through the institutional landscape, and it has acquired, in the process, a degree of apparent solidity that its underlying methodology does not really support.

This piece is the first of two on the framework that has produced this figure. The argument here is that the framework is wrong in a specific way: it incorporates pent-up demand as a stock variable requiring satisfaction, but ignores a symmetrical phenomenon — pent-up supply. When you correct for this omission, the projected housing requirement comes down. The second piece, to follow, looks at the demand side of the calculation in detail and shows that the figure is over-stated for additional reasons that compound with this one. For now, the conceptual point is enough.

This matters because the consequences of getting it wrong are not symmetric. We have been here before.

Why the standard framework cannot see the other half

The framework that produces the deficit estimate works like this. You take the observed average household size from the Census — 2.74 in 2022 — and compare it to an underlying preferred household size that would supposedly be seen if housing were adequate. The Commission’s estimate of underlying household size is around 2.4 to 2.45. The gap between observed and underlying, multiplied by the population, produces the deficit. Roughly: if Ireland had enough housing, more households would have formed, household size would be lower, and the implied additional housing requirement is the difference.

This is a coherent way of thinking about the question. It has the structure of a counterfactual: how many more dwellings would be occupied if conditions were different from how they are? The problem is not the structure. The problem is that the framework asks the counterfactual question on the demand side only.

Here is the same question applied to the other side of the market. Irish residential transaction volumes have been suppressed for over fifteen years. Long-run historical norms for housing market turnover sit somewhere between 4 and 6 per cent of stock per year, across most developed economies. Pre-crisis Ireland averaged around 4.8 per cent. Since 2010, Irish turnover has run at roughly 2.5 to 3 per cent — between forty and fifty per cent below the long-run norm, for fifteen consecutive years.

The conditions producing this suppression were a perfect storm in 2010 and the effects have been compounding ever since.

A number of factors created what is understood in the US market as the “lock-in effect”. Negative equity meant households who could not sell without crystallising a loss. The build up of long term mortgage arrears and restructuring arrangements. Ultra-low interest rates incentivised people to stay put to retain tracker mortgages. The absence of a bridging finance market prevented households from buying before selling. The collapse of new-build supply removed the trade-up options that would have unlocked the chain.

Notice that these are the same conditions that the Commission cites as producing pent-up demand. Negative equity, low new-build supply, shrinkage of intermediate stock, and so on. The conditions do not produce a one-sided phenomenon. They produce a market that has stopped clearing. When a market stops clearing, every transaction that would have happened under stable, functioning conditions but did not is a missed transaction. Some of those missed transactions would have been households forming. Some would have been households moving.

The Commission counts the first. It does not count the second. There is no analytical reason for this. It is simply a feature of how the framework was constructed.

What pent-up supply actually means

Take a household that would have moved in 2014 but did not. Perhaps they wanted to downsize but no suitable smaller dwelling existed in their area. Perhaps they wanted to trade up but could not afford the gap. Whatever the specific friction, the move did not happen. That household is now sitting in a dwelling they would, under stable conditions, have left. The dwelling itself is occupied — it is not vacant — but it is locked-in. It is not available to anyone else, even though under normal market conditions it would by now have been released two or three times over.

Multiply this by fifteen years of suppressed turnover, and the cumulative quantum is substantial. The rough arithmetic: if Ireland’s annual transaction volume should have been around 100,000 per year at long-run norms, and has actually been around 55,000 per year, the cumulative gap over fifteen years is over half a million transactions that did not occur.

That is not the same as saying 500,000 dwellings did not enter the market, because most transactions are internal chains — Person A selling to Person B who sells to Person C — and the net effect on listings at any moment is much smaller than the gross transaction count. The relevant figure is the net stock of households currently sitting in sub-optimal arrangements. That number is harder to estimate precisely, but it is unambiguously in the high tens of thousands at minimum, and plausibly substantially higher.

The chain effect

The standard objection to including pent-up supply in housing demand calculations is that internal reallocation of existing stock does not change total supply or total demand. If a downsizer moves to an apartment and sells their family home to a trader-up, the same dwellings exist before and after; the calculation is unchanged. So the argument runs.

This argument is incomplete because it treats housing allocation as a series of static bilateral matches rather than as a multi-stage system in which one unlocked transaction can release dwellings into the market that satisfy multiple unmet demands.

Consider three households. A retired couple in a four-bedroom house wants to downsize. A family with growing children in a three-bedroom semi wants to trade up. A young couple in a rented apartment wants to buy a starter home. Each of these households wants something they cannot get. Each is also sitting on something somebody else wants. In the standard framework, this scenario is read as three units of unmet demand, requiring three new dwellings to satisfy. A deficit of three.

But look at what happens if just one of these households is enabled to move. Suppose the retired couple is given access to bridging finance and downsizes. Their family home enters the market. The trading-up family buys it. The semi they vacate enters the market. The young couple buys it. One unlocked transaction has produced three completed transactions and satisfied three units of unmet demand. No new construction has occurred.

This is the chain effect, and it is well-documented internationally. UK Treasury analysis of downsizer mobility, academic work on Dutch housing chains, Australian research on retirement housing transitions — all converge on the finding that downsizer-initiated chains typically produce between two and four onward transactions before terminating. The figure varies by market structure and the depth of the secondary market, but it is consistently substantially greater than one.

The implication for Irish housing forecasting is that the supply potential of the existing stock, when activated through chain transactions, is materially larger than the supply potential the same stock represents in a static view. A framework that ignores this — which is to say, every current Irish housing forecast — systematically undercounts the supply side of the equation.

The asymmetry of consequences

The natural response is to ask: so what? Even if the targets are over-stated, the result is that we build more housing than strictly necessary, which is hardly the worst problem an Irish housing system could have. Better to over-build than under-build. The visible problem is scarcity. Surplus is theoretical.

This response is wrong in a specific and important way. Over-building and under-building are not symmetric risks.

Under-building produces continued scarcity, continued price pressure, continued political demand for policy response. The system has well-understood feedback loops. The error self-corrects through pain, but it self-corrects.

Over-building produces something quite different. Construction is heavily debt-financed. Sustained over-supply collapses prices. Collapsed prices destroy developer equity. Developers fail. Banks bear losses on construction loans. Credit availability for housing finance constricts. Supply collapses rapidly. The over-supply turns into under-supply within a single cycle, and the financial damage takes a decade or more to repair.

This is not theoretical. Ireland built approximately 90,000 dwellings in 2006 against the same kind of confident long-range demand projections that we are constructing today. Within four years it was building 8,000. The damage to the banking sector, the construction industry, the public finances, and the social fabric took fifteen years to substantially repair, and arguably is not yet fully repaired. The 2008 collapse was preceded by exactly the same kind of institutional consensus.

The country that experienced the worst property crash triggered in 2008 should be the country most vigilant about avoiding a second. Instead, it is the country least willing to discuss the possibility. Every demand projection institution in Ireland has now publicly committed to a deficit framing that does not accommodate the symmetric risk. The Department of Finance has extended the time horizon for pent-up demand to 2040, which has the structural feature of guaranteeing that any future moderation in demand can be characterised as the pent-up demand still working through, rather than as evidence that the projection was over-stated. The framework has been made effectively unfalsifiable for the next decade and a half. That is not how forecasts in a healthy epistemic system work.

The country that just spent fifteen years recovering from a housing collapse produced by one-sided demand forecasts should not be calibrating its next decade of construction against another one-sided demand forecast.

What this is and is not

This is not an argument that Ireland does not need to build more housing. It does. The deficit framing may be over-stated, but the underlying scarcity in particular regions and dwelling types is real, and a substantial increase in supply is required.

This is not a prediction of a 2008-style collapse. The conditions of 2026 are not the conditions of 2007 in important ways. Credit availability is different, demand fundamentals are different, the demographic structure is different.

This is an argument for is the recognition and inclusion of an analytical category that has been omitted. The framework can be improved by including pent-up supply. The policy response can be improved by adding an allocation-focused dimension to what has become a single-axis supply-focused debate. The intellectual basis for both of these exists in publicly available Irish research.