AI-generated editorial cityscape representing AGTP's 2025 Portugal residential market analysis; not a specific property or transaction.

AGTP MARKET INTELLIGENCE

Portugal Real Estate Market: What 2025 Transactions Reveal

A full-year look at prices, property types, marketing time and negotiation—with a closer examination of new versus used homes in Lisbon, Cascais and Setúbal.

Reporting period: Full calendar year 2025

Period
Full calendar year 2025
Published
August 21, 2026
Reading time
24 min
Source status
definitive
Last reviewed: August 21, 2026

Hero: AI-generated editorial imagery created for AGTP; it does not depict a specific property, transaction or exact location.

Area unitsArea units set to square metres

The short version

The municipal average tells us where a market begins. It does not tell us what kind of property a buyer is comparing.

In 2025, new apartments averaged approximately 46% more per square metre than used apartments in Lisbon, 84% more in Cascais and 71% more in Setúbal. They also took much longer, on average, to move from listing to transaction: 9–12 months for new apartments, compared with 4–5 months for used apartments.

Houses carried much higher total transaction values than apartments because they were substantially larger. They also showed greater average negotiation in all three focus markets. Bedroom count added another layer: T3 apartments cost more overall than T2s while averaging less per square metre.

The practical lesson is simple. “Where?” is only the first question. Buyers also need to ask: new or used, apartment or house, how large, and how far through its pricing history?

How to read this annual review

This report uses full-year 2025 residential data from Confidencial Imobiliário's SIR — Sistema de Informação Residencial.

It begins with six municipalities commonly considered by the Americans I work with: Lisbon, Cascais, Oeiras, Setúbal, Palmela and Sesimbra. It then looks more closely at Lisbon, Cascais and Setúbal, where the annual data lets us separate apartments from houses, new from used properties and selected bedroom groups.

The prices are averages, not medians. Transaction figures are CI estimates anchored to INE statistics and segmented using the SIR sales structure; they are not the raw sample size supporting every other metric.

CI's definition of a new property is broader than “never previously occupied.” It can include first-hand developer sales, certain resales within five years of construction or rehabilitation, and properties sold while still in planning or under construction. That definition matters when we interpret both price and marketing time.

For complete definitions and our approach to missing data, revisions and derived calculations, see How We Analyze Portugal's Real Estate Market.

Six markets at a glance

Exact approved 2025 values for this comparison
MarketEst. transactionsAvg. €/m²Avg. transactionAvg. GPAListing to transactionFinal discountAccumulated discount
Lisbon8,885€5,399€510,85997 m²7 months4.6%6.3%
Cascais3,198€4,928€646,446131 m²6 months5.2%7.2%
Oeiras2,834€4,343€496,817117 m²7 months4.3%5.4%
Setúbal2,350€2,604€268,435107 m²5 months4.3%5.8%
Palmela1,315€2,380€274,118122 m²6 months3.7%5.5%
Sesimbra1,176€3,112€305,668114 m²5 months4.8%6.4%

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

The broad pattern will look familiar to anyone comparing these markets. Lisbon had the highest average price per square metre. Cascais had the highest average transaction value and the largest average property among these six. Palmela recorded the lowest average €/m² and a larger average property than Lisbon, Oeiras, Setúbal or Sesimbra.

Those comparisons are useful, but they are not the most revealing part of the annual data. The deeper value comes from separating the properties inside each municipal average.

Lisbon's overall €5,399/m², for example, combines new and used apartments, houses, small properties and large ones. A buyer choosing between a used T2 and a new T3 is not really shopping the municipal average. The same is true for a Cascais buyer moving between an apartment and a house, or a Setúbal buyer deciding whether new construction justifies its price.

Why the property segment changes the story

Municipality is usually the first filter in a property search. It is rarely the last.

Property type affects average area and total purchase price. New or used condition can shift the price position and marketing timeline. Bedroom count changes how much space the buyer is purchasing, but it does not always produce a higher price per square metre.

This annual dataset gives us enough information to examine those distinctions without pretending they are controlled experiments. The averages still contain different locations, specifications, views, building qualities and states of repair. But they help us move from a generic statement—“Cascais averaged €4,928/m²”—toward more useful questions about what a buyer is actually considering.

Are new apartments more expensive than used apartments?

New and used apartments occupied very different price positions

Average transaction price per m² · apartments · Lisbon, Cascais and Setúbal · full calendar year 2025

New apartments averaged €6,997/m² versus €4,797/m² in Lisbon, €8,459/m² versus €4,605/m² in Cascais, and €4,090/m² versus €2,394/m² in Setúbal.

Source data: Confidencial Imobiliário, SIR, 2025. Calculations by The American's Guide to Portugal. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Yes. In these three markets, the difference was substantial.

Exact approved 2025 values for this comparison
MarketNew apartments €/m²Used apartments €/m²AGTP calculated difference
Lisbon€6,997€4,797approximately 46%
Cascais€8,459€4,605approximately 84%
Setúbal€4,090€2,394approximately 71%

Source data: Confidencial Imobiliário, SIR, 2025. Calculations by The American's Guide to Portugal. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

The difference was not merely a few hundred euros per square metre. In Lisbon, new apartments averaged €6,997/m², compared with €4,797/m² for used apartments. Cascais recorded €8,459/m² for new apartments and €4,605/m² for used. In Setúbal, the averages were €4,090/m² and €2,394/m².

The total transaction values were equally striking. New apartments averaged €707,052 in Lisbon, €1,082,281 in Cascais and €393,854 in Setúbal. Used apartments averaged €420,325, €461,640 and €212,283 respectively.

For an American buyer, it is tempting to describe those differences as “the new-build premium.” They do show that the two segments occupied very different price positions. But they do not isolate the value attributable solely to being new.

New projects may be concentrated in different micro-locations, offer different parking or amenity packages, use different construction standards or target a different buyer than the used inventory. The average new apartments were also larger: 102 m² versus 91 m² in Lisbon, 125 m² versus 100 m² in Cascais, and 101 m² versus 92 m² in Setúbal.

What this means: New apartments transacted at much higher average prices per square metre and much higher total prices than used apartments in all three markets.

What a buyer can do: Compare the benefits of new construction with the relevant used alternatives, then consider how much renovation, energy performance, parking, warranty, completion certainty and immediate usability matter personally.

The boundary: These are segment averages, not controlled like-for-like premiums and not measures of appreciation.

Do new apartments take longer to sell?

New apartments took longer to move from listing to transaction

Average listing-to-transaction time—not construction time and not months of available inventory · full calendar year 2025

Lisbon averaged 12 months for new and 5 for used apartments; Cascais 9 and 5; Setúbal 11 and 4.

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

They took longer to move from listing to transaction in all three focus markets.

Exact approved 2025 values for this comparison
MarketNew apartmentsUsed apartments
Lisbon12 months5 months
Cascais9 months5 months
Setúbal11 months4 months

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

The contrast was widest in Setúbal, where new apartments averaged 11 months from listing to transaction and used apartments averaged four. Lisbon's comparison was 12 months versus five; Cascais was nine versus five.

That does not necessarily mean new apartments were unpopular. CI's new category can include properties sold while still in planning or under construction. A development may also release inventory over time or spend longer in the market by design. Listing-to-transaction time is not construction-completion time, and this dataset does not tell us why each individual property took as long as it did.

The negotiation figures add context. New-apartment final discounts averaged 2.2% in Lisbon, 2.5% in Cascais and 1.4% in Setúbal. Used-apartment discounts averaged 5.3%, 4.8% and 4.2%. New apartments took longer to transact but closed closer, on average, to their last asking prices.

That combination may reflect differences in how new projects are priced and sold, the role of developers, the stage at which a buyer commits and the relative comparability of units. Those are reasonable interpretations—not conclusions proven by the aggregate data.

What this means: The new-apartment segment combined higher average prices, longer listing-to-transaction times and smaller final discounts.

What a buyer can do: Establish whether a new property is completed, under construction or being purchased off-plan, and understand which dates the quoted timeline actually represents.

The boundary: A longer marketing period does not by itself establish weak demand, construction delay or greater buyer leverage.

Are houses more expensive than apartments?

Apartment and house budgets reflected both value and area

Average transaction value and gross private area · three focus markets · full calendar year 2025

Average total transaction value

Exact average transaction values
MarketApartmentsHouses
Lisbon€495,272€1,024,077
Cascais€541,156€919,582
Setúbal€233,936€382,234

Average gross private area

Exact average gross private area
MarketApartmentsHouses
Lisbon94 m²182 m²
Cascais103 m²203 m²
Setúbal93 m²156 m²

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Houses were more expensive in total in Lisbon, Cascais and Setúbal. Their price per square metre told a more nuanced story.

Lisbon

Apartments averaged €5,388/m², €495,272 per transaction and 94 m². Houses averaged €5,753/m², €1,024,077 and 182 m².

The average Lisbon house was 88 m² larger and its transaction value was approximately 107% higher than the apartment average. Houses were relatively uncommon in the municipal transaction mix—CI estimated 282 house transactions versus 8,603 apartments—which is another reason not to treat the two segments as interchangeable.

Cascais

Apartments averaged €5,092/m², €541,156 and 103 m². Houses averaged a lower €4,502/m², but their average transaction value was €919,582 because they averaged 203 m².

The average house was 100 m² larger than the average apartment and approximately 70% more expensive overall. Cascais is a particularly clear example of why a lower €/m² can coexist with a much higher purchase price.

Setúbal

Apartments averaged €2,598/m², €233,936 and 93 m². Houses averaged a similar €2,626/m², but €382,234 per transaction and 156 m².

The difference in €/m² was small. The total transaction value was approximately 63% higher because the average house was 63 m² larger.

What this means: Moving from apartments into houses changed the likely total budget much more than a casual €/m² comparison would suggest.

What a buyer can do: Rebuild the acquisition budget when changing property type. Do not simply multiply the apartment €/m² by a desired house size and assume the result describes the available house market.

The boundary: Houses and apartments can occupy different micro-locations and condition profiles within the same municipality.

Did houses sell further below asking price?

Houses showed larger average asking-price adjustments

Final discount and accumulated discount · full calendar year 2025

Final discount compares the last asking price with the transaction. Accumulated discount includes prior revisions.

Lisbon

Final discountApartment 4.5%House 7.5%

Accumulated discountApartment 6.2%House 9.3%

Cascais

Final discountApartment 4.5%House 7.1%

Accumulated discountApartment 6.1%House 9.8%

Setúbal

Final discountApartment 3.9%House 5.8%

Accumulated discountApartment 5.2%House 7.7%

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

On average, yes, in all three focus markets.

Exact approved 2025 values for this comparison
MarketApartment final discountHouse final discountApartment accumulated discountHouse accumulated discount
Lisbon4.5%7.5%6.2%9.3%
Cascais4.5%7.1%6.1%9.8%
Setúbal3.9%5.8%5.2%7.7%

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Houses also took longer to transact in Cascais and Setúbal. Cascais houses averaged seven months compared with five for apartments. Setúbal houses averaged six months compared with four. In Lisbon, both broad segments averaged seven months.

One plausible explanation is that houses are more heterogeneous. Land, views, condition, construction quality, privacy, pools, outbuildings and legal details can make two houses with similar nominal areas difficult to compare. Larger total prices can also narrow the buyer pool. But the aggregate data does not prove why the differences occurred.

What the figures do establish is that the average path from asking price to transaction was different. In Cascais, for example, houses averaged a 7.1% final discount and a 9.8% total adjustment from accumulated discount. Apartments averaged 4.5% and 6.1%.

For buyers, this is a reason to examine house listing histories carefully. For sellers, it is a reminder that an ambitious original price may be followed by both revisions and a final negotiation.

The boundary: A larger average discount does not justify automatically subtracting 7% or 8% from the asking price of every house.

How much more did a T3 apartment cost than a T2?

Lisbon

T2 apartment

Average €/m²
€5,225
Average transaction
€436,723
Average GPA
83 m²
Listing to transaction
7 months

T3 apartment

Average €/m²
€5,115
Average transaction
€632,782
Average GPA
120 m²
Listing to transaction
8 months

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Cascais

T2 apartment

Average €/m²
€5,103
Average transaction
€472,333
Average GPA
91 m²
Listing to transaction
5 months

T3 apartment

Average €/m²
€4,651
Average transaction
€600,700
Average GPA
122 m²
Listing to transaction
5 months

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Setúbal

T2 apartment

Average €/m²
€2,565
Average transaction
€206,603
Average GPA
81 m²
Listing to transaction
4 months

T3 apartment

Average €/m²
€2,468
Average transaction
€267,103
Average GPA
108 m²
Listing to transaction
5 months

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

An extra bedroom increased average property size and total transaction value in all three markets. It did not increase average €/m².

Exact approved 2025 values for this comparison
MarketSegmentAvg. €/m²Avg. transactionAvg. GPAMonths
LisbonT2 apartment€5,225€436,72383 m²7
LisbonT3 apartment€5,115€632,782120 m²8
CascaisT2 apartment€5,103€472,33391 m²5
CascaisT3 apartment€4,651€600,700122 m²5
SetúbalT2 apartment€2,565€206,60381 m²4
SetúbalT3 apartment€2,468€267,103108 m²5

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Lisbon's average T3 was 37 m² larger and almost €196,000 more expensive than its T2 average, despite a slightly lower €/m². Cascais's T3 was 31 m² larger and about €128,000 more expensive. Setúbal's was 27 m² larger and approximately €60,500 more expensive.

This is a useful way to think about the cost of additional space. A buyer may pay much more overall for the extra bedroom and area while receiving a slightly lower average price per square metre.

The practical question is not merely whether the T3 has a lower €/m². It is whether the additional bedroom, layout and area are worth the larger total acquisition cost.

The boundary: Portuguese typology counts bedrooms, not the quality or usability of the layout. Two T3 apartments can allocate their area very differently.

Larger houses did not follow one universal price rule

Cascais

3 bedrooms or fewer

Average €/m²
€4,354
Average transaction
€689,425
Average GPA
158 m²
Final discount
5.9%

4 bedrooms or more

Average €/m²
€4,612
Average transaction
€1,088,677
Average GPA
237 m²
Final discount
7.9%

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Setúbal

3 bedrooms or fewer

Average €/m²
€2,819
Average transaction
€298,177
Average GPA
112 m²
Final discount
5.3%

4 bedrooms or more

Average €/m²
€2,363
Average transaction
€499,722
Average GPA
217 m²
Final discount
6.4%

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

The house data provides a useful warning against carrying an assumption from one market into another.

In Cascais, houses with four or more bedrooms averaged €4,612/m², €1,088,677 per transaction and 237 m². Houses with three or fewer bedrooms averaged €4,354/m², €689,425 and 158 m². The larger-house group cost more both overall and per square metre.

In Setúbal, houses with four or more bedrooms averaged €2,363/m², €499,722 and 217 m². Houses with three or fewer bedrooms averaged a higher €2,819/m², but only €298,177 and 112 m². The larger houses cost much more overall while averaging less per square metre.

Negotiation also expanded with the larger category. Final discounts averaged 7.9% for Cascais houses with four or more bedrooms, compared with 5.9% for smaller houses. In Setúbal, the comparison was 6.4% versus 5.3%.

There is no single conclusion such as “larger houses always cost less per square metre.” The relationship depends on the market and the mix of properties sold.

What this means for buyers

Decide what “new” is worth to you

New apartments occupied a much higher average price position. Compare that difference with the benefits you value and with the realistic work required by used alternatives. A used property needing renovation is not directly equivalent to a completed new home, but neither is every used property a renovation project.

Understand the timeline

If buying new, identify whether the property is complete, under construction or off-plan. Listing-to-transaction time does not tell you when construction will finish or when you can occupy the property.

Rebuild the budget when changing property type

Moving from apartments to houses can dramatically increase total price because the average property becomes much larger—even if the €/m² is similar or lower.

Compare the relevant segment

A used T2 apartment should not be valued from a municipal figure dominated by a different mix of properties. Narrow the evidence by type, size, condition and micro-location whenever possible.

Examine the complete pricing history

The larger house adjustments show why original asking price, reductions, time on the market and the last asking price should all inform an offer discussion.

What this means for sellers

The municipality average is not a shortcut to the correct asking price.

A new apartment, used apartment and house can occupy very different price and negotiation positions within the same municipality. Sellers should focus on the property segment buyers will actually compare—not simply the broad average for the location.

Larger homes do not automatically achieve a higher €/m². Houses also showed larger final discounts and accumulated discounts than apartments across the three focus markets. That makes the original pricing decision especially important.

Michael's Perspective

American buyers often begin with location: Lisbon or Cascais? Oeiras or Setúbal? That is a sensible place to start, but the next question can change the pricing story just as much.

Are we comparing a new apartment with a used one? Is the buyer shifting from a central apartment to a house with twice the space? Is the extra bedroom truly necessary, and what does it add to the total purchase price? Has a property's asking price already been revised before we arrived?

The data is most useful when it helps narrow the comparison set. A Cascais house should be compared with the most relevant Cascais houses—not treated as an oversized version of the average Cascais apartment. A new Setúbal apartment and a used Setúbal apartment may solve the same broad housing need, but they occupied very different price positions in 2025.

The goal is not to find the single market average that tells us what to do. It is to identify which average is actually relevant to the decision in front of us.

Methodology and limitations

This report uses selected full-year 2025 residential indicators from Confidencial Imobiliário's SIR — Sistema de Informação Residencial.

Results are municipality-level averages. Prices are not medians and may be affected by the size, location, condition and mix of properties transacted. CI's estimated transaction counts are anchored to INE data and segmented using the SIR sales structure; they are not the raw record count supporting each metric.

CI's new-property definition includes first-hand developer sales, certain resales within five years of construction or rehabilitation, and properties sold during planning or construction. Listing-to-transaction time is the average period between listing and transaction; it is not construction time or months of inventory. The transaction date corresponds to the CPCV when one exists, or the deed otherwise.

Final discount compares the completed transaction price with the last asking price. Accumulated discount compares the transaction price with the initial asking price and therefore incorporates prior asking-price revisions.

AGTP's calculated new-versus-used percentages compare segment averages. They do not isolate the value caused by new condition and should not be read as appreciation.

For complete definitions, our treatment of updated selling prices, missing data, geographic normalization and source revisions, see How We Analyze Portugal's Real Estate Market.

Source: Confidencial Imobiliário, SIR — Residential Information System, 2025. AGTP analysis. © IMOESTATISTICA – TODOS OS DIREITOS RESERVADOS.

Confidencial Imobiliário/Imoestatística is the source of the underlying CI/SIR content. The American's Guide to Portugal is responsible for its selection, calculations, explanations and interpretations. CI/Imoestatística is not responsible for the adequacy or accuracy of the surrounding AGTP analysis in which selected content is incorporated.

Source status: Definitive.

Updated selling-price treatment: Updated/homogenized PVA applies.

Read Methodology & Sources

Author

Michael Minson

Michael Minson is the person and author. The American's Guide to Portugal is the website and publication.

About Michael Minson