Anatomy of a lie

Part I – What is INE, and What Does It Actually Tell Us?

“Without data, you're just another person with an opinion.”
W. Edwards Deming

The Portuguese National Statistics Institute (INE) is the country’s official body responsible for collecting, analyzing, and publishing statistical data, including on the housing market. Two of its most important quarterly publications are the House Price Index (IPHab) and the Local Housing Price Statistics.

The House Price Index provides a national perspective on how transaction values have evolved over time. Since 2012, this index has been based on actual sale prices (rather than bank appraisals), using data shared by the Portuguese Tax Authority (AT). The result is a price index (base year: 2015 = 100) that tracks percentage changes over time. For example, if the index reaches 247, it suggests that, on average, housing prices have increased by 147% since 2015.

However, national averages say little about local market conditions. Estimating the value of a property in Matosinhos, for instance, isn’t helped by knowing how prices evolved in Oeiras or Angra do Heroísmo.

That’s where the Local Housing Price Statistics come in. This report provides median €/sqm values for all 308 municipalities in mainland Portugal and the autonomous regions, as well as the 24 parishes of Lisbon and the 7 parishes of Porto. These figures are based on two variables:
– the sale price (declared for tax purposes), and
– the Área Bruta Privativa — a legally defined concept (gross private area) that serves as the standard for calculating €/sqm in Portugal.

These statistics, powered by fiscal data from the AT, form the most detailed and reliable foundation for understanding residential property trends around the country. But despite their technical robustness, they are often oversimplified or misunderstood in media coverage, where headlines may blur the difference between national and local data, or confuse quarterly changes with year-on-year trends.

Part II – Apparently, in Portugal, the number of foreigners is not the relevant criterion for estimating foreign demand

“If you torture the data long enough, it will confess to anything.”
Ronald Coase

In 2022, INE began publishing statistics on housing transactions that included a new variable: the fiscal residence of the buyer. This became possible through an updated data-sharing protocol with the Tax Authority (AT), allowing INE to distinguish between individual and corporate buyers, and to identify those with a tax address outside of Portugal.

In response to an email I received from INE, regarding the accounting of the number of residential property purchases by foreigners, they state the following:

1) "The variable ‘nationality’ is not available in the information provided by the AT".

2) "However, it should be clarified that, in terms of quantifying foreign demand, the relevant criterion should effectively be tax residence and not place of birth or nationality, as the latter do not provide any indication of the country in which the purchaser resides and where they presumably obtained the income that subsequently enabled them to purchase a home in Portugal."

Although nationality is not among the variables provided by the AT, INE appears uninterested in using it, even if it were available.
This position is revealing. According to INE, fiscal residence is the criterion conceptually “relevant” for understanding foreign demand.
What is particularly surprising is that INE seems to dismiss nationality altogether as a useful category — even though it is, by definition (in any dictionary), what determines whether someone is considered (or not) foreign.

It is equally surprising that INE seems to suggest that, because it has no information about the country in which someone resides or earned the income that enabled them to buy a house in Portugal, the accounting of purchases made by these people—even if we are talking about thousands of transactions—is not relevant.

Reading the methodological documents of the Housing Price Index and Local Housing Price Statistics provides us with clear answers: the purpose of these reports is not to track the origin of the money used for house purchases in Portugal, but to understand the dynamics of the housing market and its price evolution.

Conclusion

The public institute whose mission is to “produce and disseminate, in an effective, efficient, and impartial manner, high-quality official statistical information relevant to society as a whole” considers that the number of foreign citizens purchasing housing in Portugal should not be the relevant criterion for quantifying foreign demand.

I don't know what the reader will think, but to me it seems somewhat bizarre.

Part III – What INE Said Three Years Ago About Foreign Demand

“The single biggest problem in communication is the illusion that it has taken place.”
George Bernard Shaw

In early 2022, INE published new data on housing transactions, based on variables added through an updated protocol with the Tax Authority (AT). Two figures stood out:
– Around 15% of home purchases in 2021 were made by legal entities (companies, institutions, etc.);
– Only 5.4% of transactions involved buyers with fiscal residence outside Portugal, yet these accounted for 10% of the total transaction value.

The Local Housing Price Statistics, published a month later, reinforced this pattern: in the Lisbon Metropolitan Area, the median price paid by foreign-resident buyers was more than twice that paid by domestic buyers — suggesting even greater disparity in urban markets.

The same report included a detailed Excel table (reproduced below), covering quarterly data from Q1 2019 to Q4 2021. It showed, for each period, the number and value of homes purchased by buyers with fiscal residence in Portugal, the EU, or non-EU countries.

Key figures calculated from that dataset (2019–2021):

1. Buyers with fiscal residence outside Portugal represented, on average, 5.4% of all transactions by number.

2. They accounted for 10.0% of the total transaction value.

3. The average price paid by these buyers was €293,826, nearly double that of domestic buyers (€149,644).

4. Within that group, EU-resident buyers paid an average of €244,703, while non-EU buyers paid €374,383 — a difference likely influenced in part by Golden Visa acquisitions.

These figures are often presented as a way to measure the role of foreign demand in Portugal’s housing market. But they are misleading.

On their own, they underestimate the full extent of international demand, invite superficial media coverage or selective interpretations by experts, and risk distorting the public debate around one of the country’s most pressing challenges — as the next section will show.

IV – Something Doesn’t Add Up

"Statistics are like bikinis. What they reveal is suggestive, but what they conceal is vital."
Aaron Levenstein

What made me question INE’s approach?

After years working in real estate, I receive frequent calls from people looking to buy or rent homes in Greater Lisbon — many of them foreign nationals, either directly or through local agents.

Even accounting for some personal bias (since I work in Lisbon), I’ve never believed that the figures published by INE accurately reflect the true weight of foreign demand in Portugal’s housing market.

What’s more concerning is that the media, analysts, and even policymakers seem to have accepted this framing without sufficient scrutiny. As a result, a misleading idea has taken hold: that foreigners account for only 5% to 7% of housing transactions in Portugal.

Let’s start with common sense

It is, of course, possible for a foreign national to buy a property the moment they arrive in Portugal (or even remotely), and to change their tax residence afterward. Many buy second homes or investment properties without becoming residents at all.

However, in most cases — in Portugal or anywhere else — the first step is renting. Only after living in a new city, region, or neighborhood does someone decide where to settle more permanently.

Meaning: by the time they buy a home, many foreign nationals may already be classified as Portuguese tax residents.

And now, let’s look at the numbers

There’s at least one official statistic that suggests INE’s reliance on fiscal residence likely undercounts a significant portion of foreign homebuyers.

According to the Portuguese Court of Auditors (Tribunal de Contas), there were 114,645 individuals benefiting from the Non-Habitual Resident (NHR) tax regime by the end of 2023 — an increase of over 40,000 in a single year. This growth outpaced previous years by a large margin:
+16,371(2022), +10,797(2021), +10,596(2020), +10,808(2019).
[Source: CGOE 2023, Figure 76, p.213; data from AT]

To qualify for NHR status, individuals must meet Portuguese tax residency criteria (staying at least 183 days per year), and they must either own or rent a home in Portugal.

In other words, every newly approved NHR represents at least one home being bought or rented — a direct impact on housing demand.

The scale of this impact

In 2023, there were 136,499 home sales and 94,553 new rental contracts registered in Portugal, a combined total of 231,052 housing units.
The 40,387 new NHRs in that same year represent 17.5% of that combined figure.

Put differently: one in every six homes sold or rented in Portugal may be tied to someone entering the country under this regime — a remarkable share, even if unusually high compared to prior years.

And it’s highly unlikely these individuals were spread evenly across the country. It’s far more plausible that they were concentrated in municipalities already facing housing pressure.

Is this a plausible scenario?

One might argue that these individuals had not yet moved their tax residence to Portugal at the time of purchase or rental, and thus were counted by INE as foreign-resident buyers.

Is that possible? Technically, yes.

Is it likely? Honestly, not at all.

Over the past five years (2019–2023), the annual increase in NHR beneficiaries has always exceeded the total number of buyers officially classified as having a foreign tax residence.

In 2023, the difference was especially stark: 10,391 property transactions by foreign resident buyers vs. 40,387 new NHRs.

Some foreign clients contact me urgently, looking to rent a property within days, often offering well above market rates. Why the rush? In many cases, their goal isn’t long-term housing — it’s to secure a valid rental contract to present to authorities.

Of course, these are just individual cases. But they highlight the gap between statistics and lived reality — and why both need to be examined side by side. Sometimes, the context behind the numbers is just as important as the numbers themselves.

The lingering question

INE’s current methodology appears to exclude a large subset of buyers — overwhelmingly foreign nationals — who meet the tax residency requirements because they rented or bought a home.

If only ¼ of the 40,387 new NHRs in 2023 bought a home (while the rest rented), this would already suggest a foreign-driven housing demand nearly twice as large as INE estimates — based solely on fiscal residence.

And that’s only accounting for this one regime.

How many more foreign buyers are flying under the statistical radar?

V – The Importance of Not Losing Sight of What Matters

“The greatest enemy of knowledge is not ignorance. It is the illusion of knowledge.”
Daniel J. Boorstin

In May of this year, there was a surge of speculation about who might become the next Pope. The Portuguese media gave particular attention to José Tolentino Mendonça, the Portuguese cardinal who has lived in the Vatican since 2018, and his chances of being elected.

But how should Cardinal José Tolentino Mendonça be described?

By his Portuguese nationality? Or by his tax residence in the Vatican?

Should we refer to him as “the Portuguese cardinal” — or “the cardinal whose fiscal residence is in the Vatican”?

The apparent absurdity of the question — and to be clear, I have no knowledge of the Cardinal’s actual residency details — simply serves to illustrate a larger point: INE’s insistence that “for quantifying foreign demand, the relevant criterion must be fiscal residence, not nationality,” when left without proper context, is likely to be misinterpreted in everyday discourse.

When a Brazilian, Ukrainian, or Filipino visits a property I’m selling or renting, and I give the owner feedback about how the visit went, I don’t describe those people by their fiscal residence — not least because that information likely wasn’t even shared.

The truth is, most people (including the most careful journalists) don’t tend to frame others based on their tax residency. Not the people they meet in the supermarket, nor the ones bidding on the same home, nor their favorite film directors or football players.

And if INE’s mission is to “(...) is to produce and disseminate official statistical information, promoting the coordination, development and dissemination of national statistical activity”, it should also anticipate how the data it publishes might be interpreted — and possibly misused — in public discourse.

I'm not suggesting that INE has failed in its broader responsibilities.
I’m simply pointing out that, in this specific case, it has not succeeded in preventing the misinterpretation and misrepresentation of a key piece of information.

Just as serious as not having data on a given reality, if not more so, is witnessing the dissemination of information that is not true.

And as we will see in the next chapter, this is precisely what is happening.

VI – Distracted Experts and Journalists

“The amount of energy needed to refute nonsense is an order of magnitude greater than that required to produce it.”
Alberto Brandolini

The following excerpt comes from an article published in January 2023 on the website of JLL Portugal (later removed after this series began). It was written by Patrícia Barão — at the time, head of residential at JLL, vice-president of the country’s most prominent real estate business association, and one of the most frequently quoted voices in the Portuguese media on the housing market:

“Looking at the last four years, only 6% of homes sold were purchased by foreigners (INE data). In other words, it’s the Portuguese who drive our market, accounting for 94% of sales.”

It’s simply not true that the claim “only 6% of homes sold were purchased by foreigners” is backed by INE data.

And ironically, it’s the phrase “(INE data)” — intended to lend credibility — that makes the statement demonstrably false.

Because if INE has never stated that “only 6% of homes sold were purchased by foreigners” (and it hasn’t — it has published no such figure), then no one is in a position to conclude, at least under the authority of INE, that “it is the Portuguese who make up our market, generating 94% of sales”.

Now consider this article from Lusa News Agency, dated 23 June 2025:

Home purchases by foreigners fall to the lowest level since 2021

This headline was repeated word-for-word across major Portuguese media outlets — including RTP, SIC, CNN, Sábado, and Jornal Económico.

Yet the INE report behind these articles — the one and only source for these headlines — does not contain the words “foreigner(s)” anywhere in its 14 pages. Not once.

Apparently, for much of the Portuguese press, “foreigners” and “non-residents” are interchangeable.

Is this kind of simplification acceptable in casual conversation?

Perhaps.

But is it acceptable in news coverage of official housing data?

There can be only one reasonable answer: no.

This is not a matter of semantics. These numbers concern one of the most vital dimensions of people’s lives, and one of the country’s most urgent policy challenges. Mislabeling them distorts the conversation — and the consequences are real.

And perhaps most surprising?

The figures I am going to share with you next will show you that the very same INE report — the one these headlines were based on — also included a different set of figures, drawn from the same source (the Tax Authority).

Yet those figures were entirely overlooked. By the media. By experts. By commentators.

And when that happens, something slightly unsettling occurs:

– It’s as if those numbers never existed at all.

VII – How different is 28.7% from 6.3%? (the day everyone could see the light — but most stayed in the dark)

"There are three kinds of lies: lies, damned lies, and statistics."
(usually attributed to) Mark Twain

In March 2025, Portugal’s national statistics office (INE) released its quarterly Housing Price Index report covering the final quarter of 2024. It included the usual breakdowns by type of buyer: individuals vs. legal entities, and fiscal residence in Portugal vs. abroad.

• 86.1% of homes sold in Portugal in 2024 were bought by individuals.
• 93.7% of homes were bought by buyers with fiscal residence in Portugal.
  This meant that only 6.3% of transactions were officially classified as “foreign” purchases.

But in a quiet shift, INE also released — for the first time — data based on the birthplace (naturalidade) of the buyer. This criterion was applied only to individual buyers and defines birthplace either as the actual location of birth or, where unavailable, the mother’s residence at the time.

The contrast was striking:

• Of the 134,540 homes purchased by individuals in 2024, 28.7% were bought by people born outside Portugal.
  That’s over four times the figure suggested by the fiscal residence criterion.

This new variable was only featured in a few inside pages of the March 2025 report and did not appear again in INE’s June publication — and yet, it completely changes how we might understand the role of foreign demand in the housing market.

Conclusion

Of the three criteria mentioned that are or may be correlated with foreign demand (nationality, place of birth, and tax domicile), the INE considers that, “(...) in terms of quantifying foreign demand, the relevant criterion should effectively be tax domicile and not place of birth or nationality, as the latter do not give any indication of the country in which the purchaser resides and where they presumably obtained the income that subsequently enabled them to purchase a home in Portugal.”

In all honesty, I cannot understand this justification, as I mentioned earlier, because I do not believe the purpose of these reports is to track the funds used to purchase real estate. In any case, INE does not have access to the criterion that objectively defines what is or is not foreign: nationality. In this sense, it would always be up to INE to define the substitute criterion.

The issue is that not only is the justification difficult to understand—at least as it was presented—but the resulting figure expresses a huge difference compared to the other available criterion: birthplace.

Shouldn't INE, regardless of its understanding, provide a little more context? Namely, by informing that the choice of another variable would lead to such different values?

This is precisely what it did in its March 2025 report (INE did not emphasize the difference, but anyone who had paid attention to those pages would have identified it). However, it failed to do so three months later, in June, when it published its subsequent report.

Because I cannot help but emphasize: 28.7% is very different from 6.3%. And if both could quantify the same magnitude (in this case, foreign demand), some context can and should be provided.

The chart below shows just how different the results are, depending on the criterion used.

To be precise, it should be noted that the 6.3% represents a relative weight in relation to the total number of properties sold (including legal entities), while the 28.7% refers to the relative weight considering only purchases by individuals, since place of birth only applies to individuals (a legal entity does not have a “place of birth”).

With the adjustment in the denominator (from 134,540 homes purchased by individuals to a total of 156,325 homes sold in Portugal in 2024), the weight of the number of homes purchased by citizens of foreign nationality in relation to the total number of homes is 24.7%.

Even so, this is four times the quantification of foreign demand based on nationality, when compared to the same calculation based on tax domicile/habitual residence criteria (24.7% vs. 6.3%).

VIII – Final Reflections

"The world is more willing to accept a simple lie than a complex truth."
Alexis de Tocqueville

Why does it matter whether a homebuyer’s connection to Portugal is defined by their place of birth, nationality, or fiscal residence?

Let me be clear: I hold no ill will toward foreigners — quite the opposite. My livelihood depends on them. The majority of my income comes from providing real estate services to international clients.

I’m probably in the 99th percentile of people who would personally benefit if the Non-Habitual Resident regime continued to offer 0% income tax, if there were no restrictions on short-term rentals, or if buying property still qualified for golden, silver, or platinum visas.

In my opinion — and borrowing the exact phrase used by INE in their email response to me — the “relevant criterion” for analysing foreign demand is income.

As discussed in Part III... Remember? The INE figures from 2019, 2020, and 2021 showed that the average price paid by foreign buyers was nearly double that paid by domestic residents. It’s fair to assume that a significant (though not necessarily majority) portion of foreigners have higher purchasing power.

Which means their ability to buy property is substantially different from that of most people born or raised in Portugal. As a result, a growing share of housing stock is naturally being directed toward market segments with higher financial capacity.

This isn’t a moral judgment against developers who choose to target those segments. It's a rational business decision — higher margins, less reliance on bank financing, and reduced exposure to economic cycles.

But if we underestimate the scale of foreign demand — particularly from high-income individuals — and ignore the trend toward developing products for wealthier buyers, we risk limiting the State’s ability to grasp the full scale of the challenge. And without that understanding, it becomes harder to design incentives that encourage development aimed at more affordable segments of the market.

And yet, in a country that so often invokes the constitutional right to housing, the tax system treats three very different people more or less the same: someone who already owns 17 properties, a foreign investor who pays no income tax in Portugal, and someone buying their first home to live in.

To be fair, the previous government did take a step in the right direction — a full IMT exemption (property transfer tax) up to €324,000, and a partial exemption up to €648,000, for buyers under 35.

But honestly, shouldn’t a 40, 50, or 60-year-old who’s never owned a home — and is only now able to buy — also benefit from that exemption?

Does the State really need to keep taxing the purchase of a primary residence — especially in a context where IMT revenue continues to grow year after year?

Recent failures

Let’s not forget that former Prime Minister António Costa promised 26,000 new homes to be delivered over six years. That promise was made in 2018, with a target date of 2024. But by September 2023 — just two months before his resignation — only 1,400 units had been completed (and when the next government took office, the figure had risen only slightly to 1,600). That’s a 6% execution rate.

Or that, in late 2023, Lisbon City Council (CML) was preparing to relaunch two major tenders for public housing in Benfica and Parque das Nações. These projects had failed to attract a single bid between October 2021 and December 2022 — and as of early 2024, they remained on hold due to political gridlock within the municipality.

In September 2024, the new government set a far more ambitious goal: 59,000 new homes by 2030. And in April 2025 — just before a new election campaign (which they won again) — that number was increased to 130,000 homes through construction or rehabilitation.

Given that Prime Minister Luís Montenegro has been in office for just over a year, it may be too early to demand concrete results. But one thing seems clear: these homes will only be built if conditions are created to attract private investment.

What really matters

When INE [as the producer of official data] makes a choice that potentially underestimates reality — and fails to clarify that other available variables would lead to substantially different figures — then it bears some responsibility for how that data is interpreted by experts and journalists [as opinion shapers].

And it is those experts and journalists — the ones who get media exposure — who have the greatest influence over what is commonly called public opinion.

Judging by what we’ve seen in the media, it seems neither journalists nor experts have paid much attention to detail (perhaps understandable given the demands of today’s digital content cycle). Nor do many seem particularly concerned about accuracy — at least, not when compared to their efforts to shape public opinion in ways that serve their interests.

Often, those interests are economic in nature — yet media outlets rarely disclose them. After all, how relevant would it be to let the audience know that the person advocating for a certain position might directly benefit from that position, either personally or through their clients or employer?

Maybe it’s just carelessness. Or, as people now say, a lack of awareness. But every time this happens, a little piece of transparency is lost — and it never comes back.

It is public opinion that puts the most pressure on political action, which in turn is guided by official data or, eventually, by the most consensual interpretation that the media and experts can make of it. If the causes are underestimated (and I think this article clearly shows how foreign/external demand has been undervalued), we risk that the measures outlined to mitigate them may be doomed to failure. And considering the current housing situation, that may be what we have been doing: failing.

Why do we keep failing?

There are countless people whose favorite phrase seems to be “Phenomenon X is not responsible for the housing crisis.” At the end of the day, we have a set of causes whose effect is ignored by rhetoric based on the idea that “all housing-related problems are not the fault of this phenomenon.” This is something whose truth I do not dispute, but whose relevance deserves to be treated like a punching bag.

It is a recurring caricature—obviously, problems with access to housing are not the result of a single effect—which aims to absolve any dynamic of responsibility for its contribution to a context in which so many people with jobs and salaries cannot afford to pay for a home.

Everyone has the right to have their interests, but the public also has the right to know that they exist, especially when it is a media outlet that gives them a voice.

It is a simple contribution, but one of great importance for the transparency of any public discussion.

But in all sincerity, I don't think we are at that level.

IX – A final note on rentals (and their inexplicable exclusion from housing price analysis)

"Falsehood flies, and the truth comes limping after it."
Jonathan Swift

The impact of foreign demand on the rental market

The price trajectory of homes in a given area is not defined solely by the sales market. It is also shaped by what people are willing to pay to live — or spend a season — in someone else’s home. Ultimately, the market value of a home is a multiple of the rental value, which is estimated for someone who would be willing to pay to live there.

Foreign demand significantly impacts the rental market. In theory, even if not a single foreigner bought property in Portugal, international demand for rentals could still affect selling prices.

Take, for example, a brief survey I conducted in December 2023 using the real estate platform Idealista, as part of a larger piece I wrote on housing Demand in Portugal, included in a series I called HOUSING SPECIAL.

On December 4th, 2023:

• Of the 1,018 listings for T2 apartments for rent in Lisbon, 51% (522 listings) had an asking price equal to or above €2,000 per month.
• Of the 259 listings in Cascais, 56% (146 listings) had an asking price equal to or above €2,000 per month.
• Of the 390 listings in Porto, 25% (96 listings) had an asking price equal to or above €2,000 per month.
• Of the 101 listings in Matosinhos, 27% (27 listings) had an asking price equal to or above €2,000 per month.

It seems fair to say that the vast majority of Portuguese families cannot afford a monthly rent of €2,000.

And while it's true that Lisbon and Cascais, and Porto and Matosinhos, are among the most expensive municipalities in their respective metropolitan areas — and not everyone can afford to live there — it's equally important to highlight two points:

1. These 4 municipalities, out of 308 in the country, account for 11% of Portugal’s total population;
2. And in 2020, the average gross monthly salary in the Lisbon and Porto metropolitan areas was €1,472 and €1,191, respectively.

Another telling example

In the summer of 2018, just before the inauguration of the new Nova School of Business & Economics campus in Carcavelos, the media widely covered the surge in local housing prices.

It took me less than 30 seconds to find two articles — from September and November of that year, published by Jornal de Negócios and Eco — entirely focused on the topic.

When a top-ranked international university moves to a new beachfront campus, drawing students from around the world, prices are bound to react.

There are no published data on the number of foreign buyers or renters in the immediate area, but common sense — and the experience of those working in the local rental market — suggests that these students (55% of whom, according to the Nova SBE website, come from abroad — or 70% in the master’s programs, as stated in a sponsored article) did not rush to buy apartments near the school, but instead rented them — or individual rooms.

In other words, all available indicators suggest that housing prices rose in the Carcavelos area due to foreign demand, but this pressure didn’t show up in home purchase statistics — it came from the rental market.

And when rents increase, so does the potential yield of a given property, which in turn boosts its perceived investment value.

Overlooking rentals distorts the big picture

Trying to estimate the impact of foreign or non-resident demand on housing prices — particularly in areas like the Lisbon Metropolitan Area, the city of Porto, or the Algarve — solely based on home purchases is clearly insufficient.

The rental market matters.

And it matters even more when we’re talking about a demographic group whose very nature — such as students, digital nomads, or short-term assignees — implies greater mobility and, consequently, a higher likelihood of renting.