💥 Zero Bullshit
Chapitre 01

From bricks to paper

Wannabe communism

The concept fits in one sentence: 500 savers each put in 10,000 francs, a building is bought with the money, and the rent is shared out pro rata. A management company takes care of everything. The saver cashes a cheque every quarter. The first property partnership of this kind (a “civile foncière”) was born in 1964, created by Robert Ruel. The term “pierre-papier”, bricks-and-paper, came later, coined by Guy Marty in 19801.

Louis-Gaston Pelloux took the idea further. A property developer originally from Oran, in Algeria, he created Épargne Foncière in 1966, the first SCPI to invest in offices. Not flats. Pelloux would later be nicknamed the Pope of SCPIs. His group would sell its management companies in 2001. Épargne Foncière would end up at La Française REM, with €5.2bn of capitalisation and a theoretical withdrawal delay of 14 years. But we’re not there yet.

The law of 31 December 1970 made the status official. Sociétés Civiles de Placement Immobilier, the full name behind the acronym, came under the supervision of the COB, the forerunner of the AMF, France’s financial markets regulator. The legal framework was set, with a feature few investments offer: by buying SCPI units, you don’t become an “investor”. You become a partner in a civil company. Voting rights at the general meeting, a right to information, the option of sitting on the Supervisory Board. In a UCITS fund or a French life insurance policy, you are 100% passive. Here, you can have a say. A little.

For twenty years, the market grew quietly. No noise, no scandal. SCPIs remained a niche product, sold by banks to their wealthier clients. And then the 1980s arrived. The money flowed. Property prices soared. Inflows doubled every year between 1985 and 1987. In 1990, SCPIs were worth 11.5 billion francs.

And then, the first crash. The property bubble burst. Too many offices had been built, supply exceeded demand, rents fell. At the end of 1993, 1.4 billion francs of units were waiting to be sold. Partners wanted out, nobody was buying.

Except that the yield held up. It went from 6.28% in 1992 to 5.78% in 1993. Down 50 basis points. Not a collapse. The problem wasn’t the yield. It was liquidity. Those who didn’t sell in a panic didn’t lose money.

Same story with the subprime crisis. The average yield went from 5.84% to 6.07% in 2009. Up. Stock markets lost 40%, SCPIs kept paying out as if nothing had happened. Same again with Covid: unpaid rents, deserted offices, closed shops. The average yield lost 22 basis points. 4.18% against 4.40%. Nothing.

Except that 2008 and 2020 weren’t interest rate crises. Central banks were pumping in liquidity, money stayed free. 2023-2024 is something else. A brutal rise in rates that sent the cost of capital soaring and valuations crashing. Unseen since the 1990s.

And this time, the picture changes.

Sources: ASPIM2, IEIF3

In twelve years, inflows have been multiplied by five. Capitalisation by four. Four million French people hold units4. SCPIs have become a mass-market investment.

Except that Primonial and its €35bn under management is not Corum Origin with its €3bn. Both are called SCPIs. Both display a distribution rate. Both appear on the same comparison sites. But it’s not the same product.

Record inflows: €10.2bn. And the beginning of the end.

Interest rates had been on the floor since 2015. The ECB lent at 0%, sometimes below. Money cost nothing. Property became the default investment. Livret A savings accounts at 0.5%, capital-guaranteed life insurance funds at 1.5%, SCPIs at 4.5%. Easy choice.

The sales pitch at the time was crystal clear. I heard it dozens of times: inflation is at 2%, the Livret A at 0.5%, so you’re losing 1.5% a year, so come and find happiness with SCPIs at 4.5%. That pitch has aged badly.

Inflows exploded. €4bn in 2015. €5.5bn in 2017. €8.6bn in 2019, an all-time record. And it kept going: €7.4bn in 2020 despite Covid, then €7.4bn in 2021, then €10.2bn in 2022, a new all-time record.

The problem is that an SCPI that raises money has to invest it. Incoming cash can’t just sit in a bank account. Why? Because uninvested cash dilutes the yield. If an SCPI has €100M of property generating €5M of rent, its yield is 5%. If it raises €50M more but doesn’t invest it, it has €150M of “capitalisation” but still €5M of rent. Its displayed yield drops to 3.3%. Partners grumble, inflows dry up, the vicious circle kicks in.

So managers buy. They have no choice. Except that in 2021-2022, everyone wants to buy and nobody wants to sell. Prices rise. Yields at acquisition fall. In 2022, some SCPIs bought offices in the Paris region at yields of 3.5-4%. Historically low. But that’s the price of deploying the inflows.

Everyone knew zero rates wouldn’t last forever. But nobody wanted to be the first to bet on a rise. Managers who had sat on cash and waited would have seen their yield fall and their inflows evaporate. So they bought. At the top of the cycle. At prices they would never see again.

And then the ECB started raising its key rates, in mid-2022. In fourteen months, rates went from 0% to 4.5%. Up 450 basis points.

And then the whole property equation falls apart.

A risk-free rate at 0% was the perfect deal for property. 4% rental yield against 0% on bonds: easy choice. With rates at 2-3%, the question changes: why put up with the hassle of tenants, works and vacancy risk for one extra point? Property yields have to rise to become attractive again. And since rents don’t move, it’s the price that takes the hit.

A building that generates €100,000 of rent a year and traded at €2.5M (a 4% yield) has to drop to €1.67M to show 6%. A third off.

In June 2023, the AMF took an unprecedented decision: it asked management companies for an exceptional valuation of their property as at 30 June5. Unheard of. In reality, everyone already knew. Brokers knew. Managers knew. Some CGPs had understood and were warning their clients. Others carried on selling as if nothing had happened. All that was missing were the official figures to make it official.

The results came out in the autumn. Around twenty SCPIs had to cut their unit price. Unseen since the 1990s. Primopierre went from €208 to €115. Down 45%. Laffitte Pierre from €450 to €295. Down 34%. Genepierre from €270 to €180. Down 33%. And these aren’t small, obscure vehicles. Épargne Foncière, the SCPI Pelloux created in 1966, now at La Française, has €5.2bn of capitalisation. It lost more than €1bn of value in a few months. 60 years after its creation, a theoretical withdrawal delay of 14 years. Three Amundi SCPIs alone account for €11bn, or 12% of the market, and went through two to four successive cuts.

The walk of fame of losses

Inflows collapsed in its wake. €5.7bn in 2023, or -44% on 2022. €3.5bn in 2024, another -39%. But the worst part was the withdrawal requests. At the end of 2022, €142M of units were waiting for a buyer. By the end of 2024, that figure had reached €2.4bn6. Multiplied by seventeen in two years.

In practice, this means thousands of savers want out and can’t. They asked for their units to be bought back six months, a year, two years ago. They’re still waiting. On some SCPIs, the theoretical repayment times exceed a decade. On Primovie, 1.36 million units are waiting to be bought back. At the current pace of matching (close to nil), some calculate a 130-year wait7. On Primopierre, the calculation gives 1,144 years. It’s absurd, obviously. No SCPI will survive a century in these conditions. But it gives an idea of the scale of the blockage.

As Zero Bullshit revealed a few months ago (in French), a fund manager at Praemia told its distributors last year that he saw no return to normal for ten years. And he wasn’t even talking about withdrawals.

And insurers didn’t help. When an insurer holds SCPI units in its life insurance contracts, the insurer is the partner, not the saver. And when it decides to pull €50M out of a vehicle to guarantee liquidity for its own clients, it blocks the market for everyone else. Insurers are obliged to guarantee liquidity to their policyholders. To do so, they choke the vehicles. Direct investors pick up the bill.

Except that the crisis didn’t hit everyone. 18 SCPIs raised their unit price in 2025. Iroko Zen, Transitions Europe, Remake Live, Corum Origin. Names that meant nothing to anyone five years ago. Today, they capture most of the inflows. 86% of gross inflows went to SCPIs with no units awaiting sale in 20248. The others, the ones with queues, raised almost nothing.

Two worlds.

On one side, the historic SCPIs, weighed down with Paris-region offices bought at the top of the cycle between 2018 and 2022. Big buildings, big tickets, single tenants. When the market turns, they’re left with unsellable assets and valuations in free fall.

Offices. In 2022, they captured 41% of inflows. In 2025: 14%9. A brutal fall. Three factors. Remote working, first: demand for m² per employee has structurally fallen. Obsolescence, next: older buildings no longer meet French energy rules (RT 2012, the “décret tertiaire”), and bringing them up to standard costs more than the land is worth. Vacancy, finally: 4.4 million m² empty in the Paris region in 202310. In La Défense, Paris’s business district, the vacancy rate reached 33%.

Cœur Défense is the perfect example of what went wrong. The largest office complex in Europe. 160,000 m² spread over five towers. Bought for €2.1bn by Lehman Brothers in 2007, at the height of the bubble. Lehman went bust the following year, and the building went into safeguard proceedings, a French court protection from creditors. Bought back for €1.3bn by Lone Star in 2014. Sold on for €1.8bn in 2017 to an Amundi/Praemia/Crédit Agricole Assurances consortium. Three Amundi SCPIs are exposed to it through the SCI HOLD, a property holding company: Edissimmo, Rivoli Avenir Patrimoine, Génépierre. Energy-guzzling towers, massive renovation works to finance, a melting valuation. All for a stake in a company whose accounts are never filed.

Overall result: office SCPIs lost 9.5% in valuation in 2024, against 1.1% for the market as a whole. And 70 to 73% of units awaiting sale belong to SCPIs mostly invested in offices11.

For trapped partners, the options are limited. Wait (for years). Sell on the secondary market at a discount of 15 to 35% depending on the SCPI. Or tap a redemption fund, where one exists. It’s not an official suspension, but in practice it’s impossible to get out at the displayed price.

On the other side of the market, SCPIs that don’t have these problems. Not just the youngest. Corum Origin dates from 2012, Corum XL from 2017, and they’re among the best performers. Sofidy has managed Immorente since 1988 and has almost no units awaiting sale. Atland Voisin with Épargne Pierre is holding up. Alderan, Inter Gestion, Kyaneos show TDs above 6% and stable or rising valuations.

What do they have in common? They’re not subsidiaries of banking networks.

The SCPIs suffering most have an identifiable profile: managed by subsidiaries of big banking groups (La Française = Crédit Mutuel, Amundi = Crédit Agricole, AEW = Natixis/Caisse d’Épargne, BNP Paribas REIM), sold captively through the network’s branches, concentrated on Paris-region offices. These SCPIs raised massively in 2021-2022, bought at the top, and are now suffering valuation cuts and queues12. Praemia (formerly Primonial) and Perial, although independent, suffer from the same disease: overexposure to Paris-region offices bought at the top of the cycle.

Independent managers (Corum, Sofidy, Iroko, Atland Voisin, Alderan, Norma Capital, Sogenial) had more leeway. Less pressure to deploy at all costs, bolder geographic and sector diversification strategies, lower exposure to offices. Result: they’re the ones capturing most of the inflows today.

Recent SCPIs (post-2020) benefited from favourable timing: they invested the bulk of their inflows during the crisis, at the lowest prices. Acquisition yields of 7 to 9%, when the historic SCPIs had bought at 3-4% in 2021. But it’s not just luck. It’s also a radically different management choice.

These new SCPIs bet on granularity. Instead of buying three office towers in La Défense, they buy fifty small, diversified assets: a light-industrial unit in Spain, a nursery in the Netherlands, a ground-floor shop in Ireland. No concentration on a single tenant, no dependence on a single market. And rarely “thematic” SCPIs that are 100% offices or 100% retail. Diversification in every direction, often with a majority of assets elsewhere in Europe (with a tax bonus thrown in: no French social levies on foreign income).

This granular approach makes some people snigger. A few months ago, a head of distribution at a historic manager described Iroko’s properties as chicken coops. The contempt of the big for the small. Except that the “chicken coops” show 7.5% yields, zero units awaiting sale, and are raising their unit price. While office towers lose 30% of their value and lock their partners in for years.

18 SCPIs raised their unit price in 2025 (against only 8 in 2024). Almost all of them “new generation” SCPIs:

Table of the 2025 increases

That said, the real dividing line isn’t age, but management discipline. Some independent historic SCPIs (Corum Origin, Immorente, Épargne Pierre) avoided the excesses of 2021-2022 and are holding up well. Conversely, some recent SCPIs are copying the old ones’ mistakes: offices, France, high fees. It’s not an “old vs new” market. It’s a “disciplined vs constrained” market.

On one side, 4-5% TD and partners trapped for years. On the other, 7-8% and price increases. Not the same product.

The Q3 2025 figures seem to show a stabilisation: net inflows over 9 months at €3.3bn (+33% vs 2024), units awaiting sale almost stable at €2.38bn, unit price cuts slowing (-3.6% since January). But careful: 43% of SCPIs cut their dividend in 2025 (an average cut of 12%). Offices remain under pressure. And inflows remain ultra-concentrated: Iroko Zen (~€420M), Transitions Europe (~€406M), Corum Origin (~€376M) capture more than a third of the flows between them. 5 SCPIs capture 55%, 15 SCPIs capture 75%. The 200 others share the crumbs.

And SCPIs in difficulty are caught in a vicious circle: net outflows → forced sales to repay those leaving → falling valuations → more withdrawal requests → more outflows. 93 SCPIs out of 215 have units awaiting sale. Some will never get out.

Hold my kickback

As long as the market was homogeneous, the distribution channel didn’t matter much. You could buy any SCPI from any intermediary, and the result was roughly the same.

Today, that’s no longer true.

In a market split in two, the intermediary decides what you see... and what you’ll never see. If the distributor only has an interest in selling certain SCPIs, the saver doesn’t even get access to the right side of the market, however savvy they are. Distribution is no longer a detail. It determines access.

Historically, the SCPI is a CGP product. The independent wealth adviser sees a client, presents an allocation, and puts part of it into SCPIs. A relationship of trust, personalised follow-up, advice paid for by retrocessions, the commissions the product passes back to whoever sells it. For a long time, it was the only channel.

Then the bank networks got in on it. Not to give wealth advice. To shift stock. Because, as we’ve seen, every bank has its in-house SCPIs.

But since the arrival of the internet, digital platforms have flourished. Some combine the best of both worlds: a 100% online journey for those who want to manage on their own, and a dedicated adviser for those who prefer support. Access to the market remains selective (around sixty SCPIs in general, not the whole market), but the selection is made upstream. Some management companies also let you subscribe directly, but that doesn’t change the model much.

🫰 SPONSOR: LOUVE INVEST

Louve Invest is one of these platforms. Launched in 2021, it offers around sixty selected SCPIs and passes part of its commission back to the client. That’s the cashback: 3% on eligible SCPIs. On €10,000 invested, €300 goes back to the investor.

Where does this cashback come from? From entry fees. We explain right below.

Visit Louve’s website

All these channels show the same unit price. All claim to offer “free advice”. Except that nobody works for free.

The classic model: when you buy SCPI units, you pay subscription fees. Between 8 and 12%. Of these fees, 5 to 7% go into the distributor’s pocket13. That’s the retrocession. The CGP who “advises” Épargne Foncière pockets €500 to €700 on €10,000 invested. Their advice isn’t free. It’s paid for by the SCPI, which takes it out of the investment. So the adviser is the manager’s real client, not the saver. And Louve’s cashback? It comes from there. Instead of keeping the whole retrocession, they pass part of it back.

The problem is that not all products pay the same. A CGP who places €10,000 in SCPIs pockets €500. The same amount in life insurance with 2% fees? €10014. You can see why SCPIs get “recommended” so often. All the more so as it’s much simpler than setting up a buy-to-let investment.

As for SCPIs with no subscription fees, they still pay distributors. Upfront, out of their own cash. Some also offer “run”: a recurring retrocession on management fees, every year. The distributor is never working for free, and that’s perfectly normal.

Fee-for-all

Subscription fees, so the line goes, cover the search for properties, notary fees and marketing costs. That’s what you often hear. It’s largely false. Let’s look at the breakdown in an official document. At Praemia REIM (formerly Primonial REIM), it’s crystal clear:

  • out of ~11% subscription fees including VAT;
  • more than 9% go to fund-raising fees (in other words, the distributors’ commission);
  • less than 1% to fees for researching and executing investment programmes.

As for the famous notary fees, France’s property purchase costs? They’re paid by the SCPI when it buys a building, out of the capital raised. Not by the management company out of the subscription fees.

Subscription fees are therefore essentially the distributor’s commission. The rest is storytelling.

Another amusing misconception: that these fees are only really charged when the units are sold. That’s even more false. And it’s this confusion that costs savers dearly.

The mistake comes from the “withdrawal value”. Subscription at €100, withdrawal value displayed at €90. The temptation is to conclude: “I’ll pay the difference when I sell”. Except that’s an accounting illusion, and a fundamentally false one.

The reality: subscription fees are paid every year. Because it’s money that is never invested.

The mechanism is vicious: on a €100 subscription, €10 goes in commission. Only €90 is actually invested in property. And every year, the rent is earned on €90, not on €100. The difference is lost income. Permanently. Until you exit.

Some SCPIs have chosen to scrap these subscription fees. Novaxia Neo was the pioneer in 2019. Others have followed since: Iroko Zen (2020), Remake Live (2022), Upeka (2023), Mistral Sélection (2024, launched by Swiss Life), Eden, Iroko Atlas. About ten today. They all follow the same logic: zero on entry, higher management fees, commissions on transactions.

Which brings us to the other classic argument, widely pushed by managers who gorge themselves on fees: that SCPIs with no subscription fees make up for it elsewhere. Another argument parroted by several newspapers... wrongly. Precisely because the subscription commission costs you every year.

Take two SCPIs with the same gross rental yield of 7%:

Difference: 0.31 points of extra yield every year. Doesn’t sound like much? In relative terms, that’s +5.5% performance. Not once. Every year.

And here is what might look like a paradox: the SCPI with no subscription fees has management fees 20% higher (12% vs 10%). It still performs better. Because the invested capital generates rent. And the commission is taken on the income, not on the capital, which is much larger. The capital that went in commission, on the other hand, generates nothing. Ever.

More, more, more

Subscription fees aren’t the only ones. An SCPI can charge up to six different types of fees, some of which we’ve just touched on:

Management fees pay for the management company’s work. Selecting buildings, negotiating leases, looking after tenants, works, disposals. Between 10 and 18% of rents depending on the SCPI.

The argument you often hear: it makes no difference, since the displayed yield is net of management fees, and whether the SCPI takes 10% or 18%, what counts is the final TD. True. But incomplete.

Because for the same yield, an SCPI that takes 18% had to find more gross rent than one that takes 10%. To pay the same dividend, it took more risk. Less solid tenants, less prime locations, shorter leases. The TD doesn’t tell you that. The management fees do.

Paradoxically, piling up fee lines isn’t necessarily a bad thing. It’s sometimes even a sign that interests are aligned.

Let’s take two examples.

  • Iroko Zen (Iroko). Zero subscription fees. In return: an acquisition commission of 3.6% incl. VAT, a works commission of 6% incl. VAT, a disposal commission of 5% incl. VAT (but only if the capital gain exceeds 5%), management fees of 14.4% incl. VAT.
  • Remake Live (Remake AM). Same logic. Zero entry fees. An acquisition commission of 2.5% incl. VAT, management fees of 18% incl. VAT (the highest on the market), a disposal commission of 1% incl. VAT on the sale price.

Two different structures, same philosophy: the management company makes money when it acts, not when it sells units. Iroko is betting on works and capital gains. Remake is betting on day-to-day management, with higher fees but lower acquisition commissions. If the SCPI sells without a capital gain, if it lets the buildings rot, it gets nothing. It makes money when it acts well, not when it sits still.

Conversely, an SCPI that takes 12% on entry and nothing afterwards has no incentive to manage well. Once the commission is pocketed, whatever happens to the portfolio doesn’t matter.

12% entry fees are a rentier’s business model. The CGP pockets their commission, the SDG (the management company) pockets its commission, and the investor waits ten years to get their stake back.

That said, the no-fee model isn’t the answer to every problem. An SDG may be tempted to buy and sell too often to generate acquisition commissions. High management fees (14-18% vs 10-12%) weigh on the long term. And the model is young: none of these SCPIs has been through a full property cycle. Impossible to know how they’ll behave when their portfolios need major renovation work, in ten or fifteen years.

Still, the idea is good. A fee structure that rewards action rather than sales is progress. And the performance is there: Iroko Zen at 7.32% in 2024, Remake Live at 7.50%. Hard to ignore.

Oh, and while we’re at it, a clarification is in order: “no fees” doesn’t exist. Even though plenty of articles talk about it. Management companies have to be paid. So do distributors. What exists are different fee structures. Payment comes either on entry, during management, or on exit. What changes is where and when.

Still, those who popularised the term did the industry a favour. They forced a transparency that didn’t exist. That said, as Meelton Friedman would say (in French), or a famous newsletter: there is no free lunch.

🔑 Key takeaway: the fee structure isn’t enough to judge an SCPI. Corum Origin takes 12% on entry and shows one of the best performances on the market. Primonial takes 11% and is struggling. What counts is what the management company does with the money, not how much it takes. But fees give a clue: a no-fee SCPI has to prove its worth through performance. A 12% SCPI backed by a banking network can afford to be mediocre.