A yield, in theory, is simple: what you get back divided by what you put in. Except that in the world of SCPIs, nothing is ever simple. And as elsewhere: the devil hides in the details, paved with more or less good intentions.
What goes in the numerator? The gross dividend? Net of foreign tax? With or without one-off capital gains? And in the denominator? The subscription price? The withdrawal value? The average price over the year? The price on 1 January?
In the 50 years SCPIs have existed, the sector is still working out how to measure their performance. Four methods in thirteen years. Four. Each time with debates that drag on forever at ASPIM, a regulator that is always behind, management companies dragging their feet, and in the end indicators that get changed just before the market turns. Handy.
In the beginning was the DVM, for “Distribution sur Valeur de Marché”, distribution on market value. Simple on the surface: the annual dividend divided by the average unit price over the year. Except that “simple” rarely lasts in the world of SCPIs.
The problem with the DVM was manipulation of the denominator. An SCPI that raises its unit price during the year? Its DVM mechanically falls. An SCPI in trouble whose price falls? Its DVM rises. A clever manager could therefore inflate its displayed “yield” by paying generous dividends just before a price rise. The most skilful added +0.5 points of apparent yield without changing anything in the business.
ASPIM sounded the alarm. In 2012, everything changed.
The TDVM, distribution rate on market value, arrived on 1 July 2012. Officially, it was meant to replace the notion of yield and harmonise practices. In reality, the DVM had become too easy to manipulate, and ASPIM wanted a cleaner indicator.
The calculation stays close: gross dividend (before France’s flat-rate withholding tax) over the average buyer’s price for the year. The difference? You take the average price weighted by transactions, not just the displayed price. Harder to fiddle. In theory.
For nine years, the TDVM became THE benchmark. Rankings, comparisons, sales pitches. Everything revolved around it.
Except that the SCPI market was changing. Managers started investing massively elsewhere in Europe (Germany, the Netherlands, Spain). And that’s where it gets messy. How do you compare a 100% French SCPI with an SCPI that is 60% foreign? Tax treaties, withholding taxes, tax credits... Some display their dividends gross of foreign tax, others net. Impossible to make sense of it.
And then the TDVM had a structural bias that nobody really wanted to fix: it favoured SCPIs whose unit price was falling. A healthy SCPI that revalues its portfolio? Its TDVM falls. An SCPI in trouble that never raises its prices? Its TDVM stays “stable”. The market put up with it as long as everything was going up. But that wasn’t going to last.
At the end of 2021, ASPIM announced another change, effective 1 January 2022: the TDVM gave way to the TD, the Taux de Distribution, or distribution rate.
The main difference? The denominator. The TD is calculated on the unit price on 1 January of the year, not on the annual average. Goodbye to the bias from mid-year revaluations. In theory.
The change was also driven by a less avowable reason: the property market was starting to wobble. Rates were rising. Valuations were crumbling. With the new calculation, unit price cuts during the year no longer affect the TD. Just as those cuts were about to become commonplace. A coincidence of timing.
ASPIM also introduced the RGI (Rendement Global Immobilier, overall property return), supposed to take into account changes in the value of the portfolio. But in practice, the TD alone makes the headlines. Simpler. Sells better. And above all: it doesn’t show the damage to capital when the price falls.
While ASPIM debates its indicators, the AMF watches. And it doesn’t like what it sees.
Corum AM opened the ball in 2020. The AMF’s Enforcement Committee sanctioned the management company over promotional documents it found unclear, inaccurate or misleading15. Four complaints were upheld. Highlighting the benefits at the expense of the risks. Misleading information on fees. Hard-to-understand and misleading statements on performance. Fine: €600,000. At the time, Corum managed close to €1.5bn for 24,000 unitholders. Almost all of them private individuals.
The message seemed clear: the AMF would no longer let through marketing material that dressed up reality.
Except that the sanctions kept coming. Eres Gestion got €200,000 in July 2023 for failing in its duty to disclose retrocessions16. Inter Gestion REIM followed in 2024: the company and its two executives were hit with €1.3M in combined fines for gaps in its compliance set-up17. Fiducial Gérance was singled out for product sheets with no mention of the risks and one-sided promotional videos18. Always the same complaints. Always the same practices. The fines keep falling, but the sector doesn’t seem to learn the lesson.
And then 2024 came, and it exploded. The AMF Ombudsman’s report was damning: 141 SCPI cases handled, against 86 the year before. +64% in a year19. SCPIs became the third most common reason for complaints, behind the PEA, France’s tax-advantaged share savings plan, and employee savings schemes. The complaints concerned redemption delays, the falling value of units, but above all bank advice. Savers were challenging the recommendations of their advisers. The very ones who had sold them Paris-region office SCPIs in 2021, at the top of the cycle.
The regulator sanctions. Savers complain. And ASPIM, meanwhile, carries on debating the right formula to calculate a yield. Admittedly with a bit more conviction than France Whatsit Crowdfunding, which shuts its eyes and hopes the problem goes away.
October 2025. ASPIM finally gave birth to a reform. After months of meetings, lobbying and threats to walk out, a new indicator became mandatory: the PGA, for Performance Globale Annuelle, annual overall performance20.
The calculation is disarmingly simple: the distribution rate plus the change in the unit price. That’s it. You wonder why it took 15 years to get there.
An SCPI showing a 6% TD but whose price has fallen 10%? Its PGA is -4%. No more rankings celebrating “record yields” while the capital melts away.
ASPIM added several safeguards21:
The RGI, introduced four years earlier, was quietly demoted to the rank of complementary indicator. It will have lived four years. Like its predecessors.
This change didn’t fall from the sky. It is the result of several months of work within ASPIM, carried out, it says, in close collaboration with the regulator22. What the official communication doesn’t say is that the debates were tense. Very tense.
Several of the youngest management companies threatened to leave ASPIM during discussions that several managers describe as particularly stormy. Their complaint: the new rules do correct some abuses (sponsor prices, annualisation), but they also (above all) structurally favour the historic players.
How? Through a subtle mechanism.
By putting the PGA front and centre and the IRR in the background (only an obligation to present it over a minimum of 5 years), the new official indicator wipes out the competitive advantage of no-fee SCPIs. And openly lies to investors, while it’s at it.
Poof.
The new SCPIs that had made low fees their selling point find themselves on an apparently equal footing with the banking giants charging 12% entry fees. ASPIM brings together ~120 management companies23. The historic heavyweights (bank subsidiaries, insurers) carry more weight there than the challengers. After the vote, several managers of “new SCPIs” privately expressed their bitterness. Without the final result really surprising them.
There are two possible readings:
Both are probably true at the same time. But neither really works in the partners’ favour.
Still, the PGA corrects a major bias: it finally shows the real performance received by the investor, dividends AND change in capital included. In 2023 and 2024, the RGI of many office SCPIs was catastrophic: -5.1% on average for the category in 202424. Some historic SCPIs showed double-digit negative overall performance. With the PGA, these figures will now be front and centre. Impossible to hide.
The lovely numbers of the SCPIs of La Française
Except that the PGA arrives... after the worst of the crisis. The most violent price cuts happened in 2023 and 2024. The PGA will only be calculated from 2025. The -15% or -20% of some SCPIs over 2023-2024? They will never appear in an official PGA. They will stay in the limbo of the RGI, an indicator nobody looks at any more.
And above all: the PGA doesn’t fix everything. Because over the years, management companies have built up an arsenal of techniques to inflate the displayed yield. Some are now banned by the reform. Others remain perfectly legal. That’s actually partly why the PGA exists: too many shop-window biases had made the TD unreadable.
Four methods in thirteen years. And each time, the change comes just as the old method starts showing awkward figures.
🔑 Key takeaway: the battle over indicators is also a political battle. The PGA suits the big SCPIs (it hides subscription fees). The end of annualised TDs penalises young SCPIs. Every reform has its winners and losers. The investor, for their part, always has to read between the lines.
Because calculation methods have indeed been cheerfully used for years to show certain SCPIs in a better light. But these “biases” aren’t scams. Most are perfectly legal. And they have been used as much by the historic players to hide cuts or mistakes as by young SCPIs trying to exist in a locked market.
But the market is an oligopoly. The ten biggest management companies hold more than 60% of assets. They have the networks, the budgets, the name recognition. Launching an SCPI in 2023 with zero track record against Immorente or Édissimmo, which have been around for 30 years? Tricky.
So you play with the rules. You use sponsor units to reward the first investors. You communicate an annualised TD because you only have 6 months of history. You ride the accretive effect because inflows are taking off. That’s the game.
The problem isn’t that these techniques exist. The problem is that they aren’t explained. The average investor sees a ranking with Comète at 11% and Édissimmo at 4.5%. They don’t know these figures don’t measure the same thing. And nobody tells them.
So here are the six main techniques. Three are now banned or regulated by the PGA reform. Three remain perfectly legal. The aim here isn’t to say they’re wrong, but to give you the keys to compare like with like.
This was the most common bias. An SCPI launched in June pays dividends over six months. The marketing reflex: “annualise” to show a TD over 12 months. If the SCPI paid 3% over six months, you announce 6% annualised. Except that nothing guarantees the next six months will be as good. An SCPI’s first months are often boosted by the accretive effect (we’ll get to it). Annualising means extrapolating a temporary performance as if it were permanent.
And yet ASPIM had been clear since February 202325:
In substance: a fund created during the year cannot annualise its distribution rate.
So had the AMF26:
In substance: no communication is allowed on the performance of an SCPI with less than 12 months of history.
That didn’t stop many. Reason shows an annualised TD of 12.60%. Darwin RE01 announces 12.36%. Figures extrapolated from only a few months of activity27. Wemo One announces a TD equivalent to 10%... calculated on 6 months of dividends28. Upêka, launched in September 2023, was projecting an annualised TD of 7.45% from its first months29.
Some comparison sites actually did the teaching job the industry should have done. Louve Invest in particular created two separate rankings: official TDs and annualised TDs, with clear warnings on what each figure means30. The IEIF, for its part, mixed the two in its official rankings, in agreement with the management company, it said31. Handy.
Impact on investors: +1 to +3 points of fictitious TD depending on the extrapolation period
Status: ❌ Banned since October 2025 (PGA reform)
When an SCPI launches, the first investors buy at a reduced price. A discount of 5 to 10% on the normal price. The official idea: reward the early adopters who take a risk on a product with no track record. It’s actually rather cool.
The problem is that the TD is calculated on the unit price on 1 January. If that price is the sponsor price (lower), the TD is mechanically inflated.
Take Comète, Alderan’s SCPI launched in December 2023. The sponsor period lasted 2.5 months (from 18 December 2023 to 29 February 2024). Sponsor price: €237.50. Statutory price: €250. The 2024 dividend was €26.56 per unit.
The calculation is simple:
Half a point of TD that doesn’t exist. The official ASPIM TD in fact uses 10.62%32. But Comète communicated 11.18% in its first rankings33. Some distributors chose to show only the real TD, like this CGP, who writes:
In substance: they prefer to communicate 10.62% rather than 11.18%, which they consider misleading.
The Comète case also shows you can play the game intelligently. Alderan limited the sponsor period to 2.5 months and documented it clearly. Not every player is as transparent. Osmo Énergie announced a TD above 9%... calculated on a sponsor unit price of €150 against €300 at the normal price. Iroko Atlas in September 2025 raised €20M in a few hours on a sponsor offer with a 9% discount34.
And the “best SCPI” rankings? They cheerfully mix TDs on sponsor units with normal TDs. Without distinction.
Impact on investors: +0.5 to +1 point of inflated TD depending on the sponsor discount
Status: ❌ Banned since October 2025 (PGA reform)
This is THE mechanism that explains why young SCPIs show TDs of 8 to 11% while the old ones top out at 4-5%. And it’s the most subtle.
When you subscribe, you have to wait 3 to 6 months before receiving your first dividends. That’s the vesting period. Meanwhile, the SCPI invests the money and collects rent. But it isn’t paying the new partners yet. The rent is split between fewer partners. The TD is mechanically boosted.
Iroko published the calculations for Iroko Zen in 202435. The SCPI’s vesting period is 3 months. The actual 2024 TD: 7.32%.
With a 5-month vesting period? The TD would have risen to 8.08%. With 6 months? 8.49%.
That’s +1.17 points of fictitious TD just by lengthening the period by 3 months. Without changing anything in the portfolio or the rents36.
For an SCPI with a 5-month vesting period (the market average according to Corum37), the accretive effect is therefore worth about one extra point of TD compared with a transparent SCPI like Iroko Zen.
The circle is virtuous (for the SCPI): good displayed TD → attracts investors → new inflows → new accretive effect → TD stays high. And so on. As long as inflows are strong, the TD holds.
Except that the accretive effect is temporary. When inflows slow, the TD “normalises” towards the portfolio’s real yield. Investors who bought on the strength of a 10% TD discover a TD of 6-7%. No lie. Just a mechanism nobody explained to them.
The counter-move: look at the acquisition yield (AEM) published by the SCPI. That’s the real yield of the buildings, excluding the accretive effect. If the displayed TD is 10% and the AEM 6.5%, 3.5 points come from the accretive effect.
Impact on investors: +1 to +3 points of temporary TD as long as inflows are strong
Status: ⚠️ Regulated since October 2025 (PGA mandatory alongside the TD)
The RAN is the Report À Nouveau, retained earnings carried forward. The reserve of dividends left undistributed in previous years. A kind of safety cushion.
If current income falls (less rent, vacant space, works), the SCPI can dip into the RAN to maintain the dividend. The investor sees a stable dividend. They don’t see that this dividend is no longer covered by rents.
Édissimmo in 2022: the RAN was trimmed to fund the payout38. The dividend was maintained. But the SCPI was drawing on its reserves. It’s like an employee raiding their savings to keep up their lifestyle after a pay cut. It works for a while. When the RAN runs dry, the dividend drops all at once. Cliff edge.
The counter-move: compare the dividend paid with current income in the annual report. If the dividend exceeds income, the SCPI is drawing on its reserves. A shrinking RAN is the signal.
Impact: Artificially stable dividend, then a sharp drop
Status: ✅ Still legal (not covered by the PGA reform)
The TD includes all income distributed. Including exceptional payouts: capital gains on building sales, non-recurring income, adjustments.
An SCPI sells a building at a nice capital gain and distributes part of it to partners. Great. Except that this capital gain inflates the year’s TD. And it can’t be repeated the following year.
A few recent examples:
The problem: these exceptional dividends are included in the official TD. An investor looking at the ranking sees a 5.5% TD and thinks it’s recurring. Except that 0.5 to 1 point may come from one-off capital gains. The following year, back to normal.
The counter-move: read the quarterly bulletin. Serious management companies separate the “recurring” dividend from the “exceptional” one.
Impact: TD inflated by 0.5 to 2 points in the year of the sale
Status: ✅ Still legal (not covered by the PGA reform)
The most vicious one. To keep it simple, the TD is calculated by dividing the dividend by the unit price on 1 January. So if the price falls, the following year’s TD mechanically rises. Same dividend, better apparent yield.
Take Primopierre, Praemia REIM’s office SCPI. The textbook case.
In 2023:
In 2024:
The TD fell by 0.58 points. But that’s misleading: without the unit price cut, with the 2024 dividend and the 2023 price, the TD would have been 3.07%. The price cut therefore “offset” 0.47 points of TD39.
In other words: the dividend fell by a quarter, but the TD only fell by 14%. The formula hides reality.
And Primopierre isn’t an isolated case. The mechanism is automatic: the TD is calculated by dividing the dividend by the price on 1 January. An SCPI that cuts its price at the end of 2023 and pays the same dividend in 2024 sees its TD rise automatically. The numerator is stable, the denominator falls. Result: some SCPIs whose price has fallen show a higher TD in 2024 than in 2023. The indicator hides the destruction of value.
Of the 34 SCPIs that have cut their price since 202340, only one (Eurovalys, 5%) showed a TD above the market average (4.72%). The others? Between 3% and 4.5%. The “stable” TD hides the destruction of value.
43% of SCPIs cut their dividends in 202441. A TD that “holds up” at 5% can hide a 20% loss of capital and a 15% cut in the dividend.
Impact on investors: TD artificially stable or rising despite a capital loss
Status: ⚠️ Regulated since October 2025 (the PGA includes the change in unit price)
Of the six biases, three are now banned (sponsor units, annualisation) or regulated (the accretive effect, through the PGA warning). Three remain perfectly legal: the RAN, exceptional payouts, the unit price effect.
Does that mean you should run away from young SCPIs? No. Quite the opposite. They are often the most dynamic, the ones buying at the right time, the ones with the lowest fees, the ones performing best. The biases we’ve just described aren’t scams. They’re mechanical effects you need to understand so as not to build false hopes.
An SCPI with a 10% annualised TD won’t pay 10% forever. But it may very well pay a steady 7% for 10 years, which is still excellent. What matters is knowing what you’re buying.
As for comparison sites, they are not all equal. Some, like Louve Invest, do real educational work with clear warnings. Others are business introducers in disguise, taking a commission on every subscription. You can tell the difference by how transparent they are about the limits of the indicators.
🔑 Key takeaway: shop-window biases are the challengers’ weapons. Young SCPIs need them to exist against the giants. The giants don’t have that problem: they have the name recognition, the networks, the inertia. Their problem lies elsewhere. It’s real performance.
7% TD in France, 7% TD in Germany. Same displayed yield. Except that at the end of the day, it’s €52.80 against €70 for €100 of rent. A 17-point gap. No ranking shows that.
On €100 of French rent, the partner pays income tax at their marginal rate (say 30%) plus French social levies (17.2%). Total: €47.20 for the taxman, €52.80 in your pocket. On €100 of German rent, the mechanism is different. Germany withholds 15.8% at source. France exempts this income from income tax (it has already been taxed) but includes it in the calculation of the average rate. Social levies? Not due on foreign-source income. That makes about €30 for the taxmen (German and French combined), €70 in your pocket.
Same displayed TD. A 17-point gap in net yield. On €100,000 invested at 7%, that’s €1,200 more a year. Every year. In your pocket, not the taxman’s.
Bilateral tax treaties explain this gap. France has signed one with every European country to avoid double taxation. Two methods coexist. The effective rate method (Germany, Spain, the Netherlands, Italy, Ireland) is the most favourable: foreign income is exempt from French income tax but taken into account to calculate the average tax rate. The tax credit method (Belgium, Portugal, the UK) is less advantageous: the investor declares everything in France, calculates the theoretical tax, then deducts what they paid abroad. But the social levies remain due.
What the indicators don’t show. The TD measures gross income. So does the IRR. So does the PGA. They were designed for a domestic market where all rents were taxed the same way. A European SCPI with a 6% TD can deliver more, net, than a French SCPI at 7%. The rankings don’t tell you that.
European SCPIs captured 74% of gross inflows in 202442. It’s not a fad. Investors have worked out where the net yield is. Corum Origin, Iroko Zen, Remake Live, Novaxia Neo: the SCPIs hogging the top of the rankings are all mostly (or largely) invested outside France.
The advantage varies with each person’s situation. The higher the marginal tax rate, the wider the gap. At an 11% marginal rate, the gain is moderate. At 45%, it can exceed 15 points. What counts: the SCPI’s geographic breakdown (available in every annual report), the specific countries (they don’t all have the same regime), and the TD “net of foreign tax” when it’s published.
That said, Europe isn’t a miracle solution. Currency risk exists for countries outside the eurozone (the UK in particular). Local property markets have their own cycles. And tax rules can change. But as things stand, the advantage is real.
🔑 Key takeaway: the advantage of European SCPIs is first and foremost a tax one. No French property income, reduced social levies. At a high marginal tax rate, the net yield gap can exceed 15 points.
For years, the IRR was the benchmark. The internal rate of return (TRI in French) is the formula that takes in everything: rents received, the change in unit price, subscription fees. Over 5 years or 10 years, it’s what you would really have earned by investing back then. Not what the SCPI distributed. What you would have pocketed, net of everything.
ASPIM published it on the front page of its reports. Comparison sites displayed it in big letters. CGPs used it to sell the dream, pointing clients to 8% a year over 10 years as proof it was rock solid.
Then 2023 came along. Unit price cuts blew the IRRs apart. An SCPI showing a 7% 10-year IRR found itself at 2%. Or even negative. The 10-year IRR of an SCPI that has lost 30% of its value in two years is a disaster. Even if its TD stays at 5%. Even if its PGA shows +2% thanks to a recent rebound. The IRR includes all ten years. Including the years of capital loss. No make-up possible.
Since then, the IRR has all but disappeared from communications. ASPIM no longer highlights it. Comparison sites show it in small print, when they show it at all. The PGA is preferred: more flattering, easier to manipulate.
For young SCPIs, a long-term IRR makes no sense anyway. No track record, no 5-year IRR. Back to the original problem: there’s nothing to measure.
The IRR has its limits too. A 10-year IRR can hide a recent deterioration: an SCPI that boomed from 2012 to 2020 and has been collapsing since still shows a decent IRR. You have to cross-check with the 5-year IRR, and the 3-year IRR when it exists. Look at the trend, not the isolated figure. And the calculation method varies: some SCPIs include subscription fees, others don’t. It’s in the small print.
Still, it’s the only indicator that really includes fees. The only one that shows what you would have earned, not what the SCPI claims to have distributed.
TD, PGA, IRR. We’ve just spent three sections explaining their limits. The temptation would be to look for THE miracle indicator, the one that tells you everything. It doesn’t exist. No figure certifies that an SCPI is “good”. Short of creating a Zero Bullshit label43.
But there are other metrics, less talked about, that complete the picture.
Liquidity isn’t a problem for getting in. An SCPI is an investment for 8 to 10 years minimum. Whether liquidity is tight at any given moment doesn’t matter when you subscribe. On the other hand, it’s an indicator of confidence. An SCPI with 10% of its units waiting to be withdrawn is an SCPI its partners want to leave. That says something about how healthy the vehicle is perceived to be. Iroko Zen and Corum Origin have almost no units awaiting sale. Primopierre has €240M, or 11% of its capitalisation.
The TOF44, the financial occupancy rate, measures the rent actually collected against the theoretical rent if everything were let at market price. Above 90% is fine. Below 85% is a warning sign. The TOP45, the physical occupancy rate, is less reliable: a building can be physically occupied but with rent-free periods. The TOF captures both.
The VNR46, the reconstitution value, represents what it would cost to rebuild the identical portfolio: value of the assets + notary fees + registration duties. The unit price / VNR ratio tells you whether you’re paying too much. Between 95% and 105% is neutral. Above 105%, you’re overpaying. Below 95%, either it’s an opportunity, or the market knows something.
The RAN47 is the reserve of undistributed dividends. It’s the safety cushion that smooths payouts when rents fall. Above 30 days of distribution is comfortable. An SCPI that distributes 100% of its income every year has no room for manoeuvre.
The LTV48 measures debt: debt / value of the portfolio. Below 25% is conservative. Between 25% and 35% is standard. Above 40%, watch out. A high LTV amplifies everything: gains when prices rise, losses when they fall. The SCPIs that suffered in 2023-2024 often had an LTV > 35%.
None of these indicators is enough on its own. An SCPI can show a 7% TD and be sinking (it’s raiding its RAN). Another can have a 5% TD and be solid (undervalued portfolio, TOF at 98%, zero debt). The trap is looking at a single figure. The work is cross-checking.
A quick example. Primopierre at the end of 2022: a decent TD (4.12%), but an LTV of 32%, a falling TOF (89%), and 100% Paris-region offices. Taken individually, no indicator was screaming disaster. Together, they painted a fragile profile in the face of rising rates. What followed proved the weak signals right. With apologies to the boo-hoo-who-could-have-predicted-it CGPs.
💡 In short: there is no ONE magic indicator. But a few major KPIs (TD, liquidity, TOF, geographic exposure) and a few secondary ones (RAN, LTV, VNR) let you form a view. The rest is detail or marketing.