FII Market

Real-estate funds are bought for the monthly distribution, so the question that matters is whether the distribution is earned. This page reads the sector's dividend yield, then measures declared payouts against the funds' own books, then looks at the buildings underneath.

Two limits.

  • A payout above what the book yields is not evidence of anything on its own — it is consistent with realised gains, return of capital and amortisation. The numbers sit side by side; no verdict is asserted.
  • Property detail is partial: CVM leaves vacancy and delinquency blank for a large minority of buildings, so those columns are reported against their own denominator rather than averaged.

FII vs FIAGRO Net Assets

Two families on the same axis over 24 months. FIAGRO's monthly file only begins 2025-05, so its line starts there — the earlier absence is the dataset not existing, not a fund count of zero.

Loading...

Yield Distribution Across FIIs — p10 to p90

The spread, not the average. A widening gap between p10 and p90 means the sector is separating into funds that distribute and funds that do not, which a median alone conceals.

Loading...

Top 25 FIIs by Dividend Yield

Funds with more than R$50mm in net assets, ranked on CVM's reported pct_dividend_yield_mes. A very high yield on this measure is a reason to read the payout-coverage table further down, not a conclusion on its own.

No Results

Which FII Filings Are Populated

cvm_fii_mensal carries three ingested subtypes. Only complemento fed the charts above; the payout section below reads ativo_passivo and geral. This table is the ground truth for what is present, so a blank payout column can be read as "not filed / not mapped" rather than "no distribution".

No Results

Payout Coverage

rendimentos_distribuir — declared distributions — against the fund's own book: net assets, quotas outstanding, book value per quota. A fund distributing far more than its book yields is paying out of capital or realised gains; the numbers are shown, the verdict is not. The n_funds_* columns track how many funds reported each field, so a fall in the sector total can be told from a fall in coverage.

Loading...
Loading...
No Results

Fund by fund below, latest month. Payout Yield on Book is the monthly payout per quota over the book value per quota (vl_patrimonial_cotas). Reported DY is CVM's own pct_dividend_yield_mes, the measure the top-25 table above uses. The two are shown side by side rather than reconciled, because they come from different filings — a large divergence between them is itself worth checking against the fund's report.

No Results

Property Explorer

Property Rows (All Periods)

136,201

Funds in Register

864

Latest Reference Date

2026-06-30

…with an Invested Share

51,145

…with Vacancy Reported

77,731

…with Delinquency Reported

77,007

Coverage first. From cvm_fii_imovel, the per-building register inside the INF_TRIMESTRAL zip. Until recently the fetcher pulled the wrong member — the alienação file, listing buildings being sold — so this table described disposals rather than holdings. CVM leaves vacancy and delinquency blank for a large minority of rows. The counters cover all filing periods and count source rows, not distinct physical buildings; the latest reference date shows how recent the register is. Blank field values are missing observations, not zero.

The share is of invested assets, not net assets, and CVM publishes the pr_* fields without a documented scale. The explorer shows source units; a threshold in those units does not establish an economic concentration percentage. Concentration in the investor base instead is the captive-vehicle screen on Suspicious Deal Screens.

No Results