)
Impact measurement in affordable housing is hard to verify because the frameworks are mature while the data flowing into them is self-reported, delayed and defined differently by every operator: in a GIIN survey published in January 2020, 92 percent of impact investors found collecting quality data a challenge. A shared transaction record can make the financial layer, meaning rent collected, distributions, costs and the conditions attached to capital, verifiable rather than asserted. It cannot measure human outcomes, and tenants' personal data does not belong on a ledger at all.
Every investment committee that funds affordable housing asks for the same things: evidence that the capital created the impact it was committed to, a method consistent enough to compare one programme with another, and a way to show donors, regulators and beneficiaries that the money did what it was meant to do. These are minimum requests for a responsible allocator. What a programme officer usually receives is a stack of project-level reports, produced on different schedules, in different formats, by operators who have every reason to present their work favourably.
Is the problem the frameworks or the data?
The frameworks are not the weak point. The field has spent more than a decade building them. IRIS+, the metrics system of the Global Impact Investing Network, was launched in May 2019 (GIIN). The Impact Management Project established five dimensions of impact, namely what, who, how much, contribution and risk, before concluding its planned five-year term at the end of 2021 and handing its resources to Impact Frontiers (Impact Frontiers, November 2021). The Operating Principles for Impact Management were launched in April 2019 by 60 investors holding more than 350 billion dollars of assets invested for impact (IFC, April 2019).
Those frameworks tell an investor what to measure and how to weigh it. They cannot generate the underlying data, verify it, or make it comparable across operators who define it differently. The investors using them say as much. In a GIIN survey of 278 impact investors, published in January 2020, large majorities described the data itself as the obstacle:
Source: GIIN, The State of Impact Measurement and Management Practice, second edition (January 2020), survey of 278 impact investors. Share rating each a significant or moderate challenge; the market-transparency item is a market-level challenge, the others organisation-level · January 2020
| Item | Value |
|---|---|
| Collecting quality data | 92% |
| Lack of transparency on impact performance across the market | 89% |
| Comparing impact results with market performance | 84% |
| Aggregating, analysing or interpreting data | 74% |
The effort is not trivial either. The same survey estimated that impact investors spend about 12 percent of their organisation's total budget on impact measurement and management, and that it takes about a quarter of impact investing staff time (GIIN, January 2020). Those figures cover impact investors generally rather than affordable housing specifically, and the GIIN notes that some respondents found it hard to separate the budget, but they make the point: a large share of the cost of impact investing is the cost of finding out what happened.
Source: GIIN, The State of Impact Measurement and Management Practice, second edition (January 2020). Averages across 278 impact investors, all sectors; the GIIN notes some respondents found it difficult to isolate an impact-specific budget. Shown as figures rather than bars because the two shares are of different wholes
| Share of total organisational budget spent on impact measurement and management | about 12 percent |
| Share of impact investing staff time | about 25 percent |
Why does affordable housing reporting stay self-reported and hard to compare?
Affordable housing is not short of data. Occupancy, rent collected, arrears, turnover, maintenance spending and compliance with eligibility thresholds are generated continuously by the ordinary running of a building. The difficulty is where that data lives and who controls it. It sits in each operator's accounting system and spreadsheets, is formatted to that operator's processes, and is transmitted when the operator compiles a report.
A foundation funding dozens of operators across several countries therefore receives dozens of reports built on different definitions of "affordable", different treatments of vacancy and different ways of allocating costs. Turning them into one portfolio view requires an analyst to normalise definitions, align periods and decide which figures to trust when they conflict. The result is expensive, already out of date by the time it is read, and still dependent on what each operator chose to send.
Sector-level standards help with the definitions. The UK's Sustainability Reporting Standard for Social Housing, launched in November 2020, gives housing providers a common set of criteria to report against (NatWest, November 2020). Common definitions make reports comparable. They do not make them verifiable: the figures are still compiled and submitted by the party being assessed.
What does European regulation actually require?
Less than is often assumed about verification, and more about disclosure. The Sustainable Finance Disclosure Regulation is, in the European Commission's own words when answering questions from the European supervisory authorities, "a transparency regulation" (European Commission, answers to questions on the interpretation of SFDR, April 2023). A product disclosing under Article 9 has sustainable investment as its objective, and the regulation defines sustainable investment to include investment in activities with a social objective, such as tackling inequality or investing in economically or socially disadvantaged communities, provided it does no significant harm to other objectives and the investee follows good governance practices (Regulation (EU) 2019/2088, Article 2(17)). The technical standards require managers to explain how indicators of adverse impact were taken into account when testing for significant harm (Delegated Regulation (EU) 2022/1288). None of this obliges a fund to commission an audit of its operators' data.
The regime is also changing. The European Commission acknowledged in November 2025 that SFDR "has effectively been used as a de facto labelling system", and proposed three voluntary product categories, "sustainable", "transition" and "ESG basics", each requiring 70 percent of the portfolio to support the chosen strategy (European Commission, November 2025). The Council adopted its negotiating mandate in June 2026 and the European Parliament in October 2026; as of mid-October 2026 the proposal remained subject to negotiation and had not been adopted (Arthur Cox, October 2026). Whatever the final categories look like, a fund that presents itself as pursuing a social objective in housing will need evidence it can stand behind, and that is a data problem before it is a reporting one.
What can a shared transaction record actually verify?
When a property is held in a dedicated vehicle and ownership of that vehicle is recorded on a shared register, the financial events that pass through the vehicle can be recorded on the same register as they happen: rent received for each unit, distributions made and when, fees charged, and payments for maintenance. Each entry is timestamped and cannot be quietly revised afterwards, and every permitted party reads the same record at the same moment. A programme officer, an auditor or a lender does not need to ask the operator for a report on those figures; the figures are the record.
Some conditions attached to capital can then be checked continuously rather than annually. If a structure commits to keeping rents below a defined cap, each recorded rent can be tested against the cap as it is collected, and a breach appears when it happens rather than in the next annual review. If distributions to investors are made conditional on that test, the condition is enforced by the structure rather than asserted in a policy document.
Two limits apply. First, the record is only as accurate as what enters it. Rent is still paid in euros through a bank account; somebody, or some reconciled process, still has to post it to the register, and that point of entry is where controls and audit belong. Second, a test can only run on data the register holds, which brings the most important constraint into view.
- Verifiable from the record itself rent collected per unit, distributions and their timing, fees and costs paid by the vehicle
- Verifiable only as an attestation tenant eligibility and rent relative to local income thresholds, checked off the ledger by an accountable party
- Not verifiable from the record housing quality, stability and long-term outcomes for households
- Must stay off the ledger tenants' personal data, including income and household composition
Source: Framework defined in this article. The last row follows CNIL guidance on blockchain and personal data (September 2018) and EDPB Guidelines 02/2025, version 2.0 (July 2026)
What must never be put on a ledger?
Tenants' personal data. Eligibility for affordable housing depends on household income and composition, which are personal data, and a shared register that many parties can read is the wrong place for them. France's data protection authority recommended in 2018 that personal data be processed outside the blockchain, with anything recorded on it limited to a commitment, a keyed hash or a ciphertext (CNIL, September 2018). The European Data Protection Board has since gone further. Its guidelines on blockchain, adopted in final form in July 2026, state that "in general, it is not advisable to store personal data on the blockchain", note that encrypted or hashed personal data is still personal data, and recommend that controllers store any additional personal data off-chain (EDPB, Guidelines 02/2025, version 2.0, July 2026).
The workable design follows from that. An accountable party checks eligibility off the ledger, under data protection law, and the register records that the check was carried out and passed, in a form that does not identify the tenant and subject to a data protection impact assessment, not the information it relied on. A funder can then verify that every unit was let after a completed eligibility check without anyone being able to read a tenant's income. Any proposal that describes portfolio-wide rent-to-income ratios being calculated by anyone from public data has not taken this constraint seriously.
What does a shared record still not measure?
The outcomes that matter most to most foundations. Whether a family that gained stable housing kept its children in one school. Whether a community that retained an ownership stake saw less displacement over a decade. Whether wealth built through a rent-to-own arrangement survived the next financial shock. None of this can be read from a transaction record, and the frameworks built to think about it, IRIS+ and the five dimensions among them, remain necessary.
The case for a shared record is therefore narrower than it is usually made, and stronger for being narrow. It resolves the first-order problem, reliable and comparable financial and operational data across operators and periods, at the point where that data is created. The resources currently spent reconstructing that layer from operator reports can then go to the harder questions that only fieldwork and follow-up with households can answer.
The bottom line
Impact measurement in affordable housing has a data problem dressed up as a methodology problem. The frameworks are sound; the inputs are self-reported, inconsistent and expensive to check. A shared transaction record can turn the financial layer of a housing portfolio into something every permitted party reads directly, and can enforce some conditions continuously rather than auditing them after the fact. It does not reach human outcomes, it depends on disciplined entry of off-ledger payments, and it must be designed so that tenants' personal data never touches the ledger. Built that way, it gives the frameworks data worthy of them. Built carelessly, it creates a new compliance problem in the course of solving an old one.
This article is part of DeReal Perspectives on real-estate access, liquidity, and tokenization. It is analysis of the market, not investment advice or an offer of any kind.
Sources
GIIN, The State of Impact Measurement and Management Practice, second edition (Jan 2020): thegiin.org
GIIN, IRIS+ history: iris.thegiin.org
Impact Frontiers, Impact Frontiers to incorporate the resources facilitated by the Impact Management Project (Nov 2021): impactfrontiers.org
IFC, 60 Investors Commit to Manage over $350 Billion in Assets in Line with New Impact Principles (Apr 2019): ifc.org
NatWest, Social housing sector launches Sustainability Reporting Standard (Nov 2020): natwest.com
Regulation (EU) 2019/2088 on sustainability-related disclosures in the financial services sector (SFDR): eur-lex.europa.eu
Commission Delegated Regulation (EU) 2022/1288 (SFDR technical standards): eur-lex.europa.eu
European Commission, Answers to questions on the interpretation of Regulation (EU) 2019/2088 submitted by the European supervisory authorities (Apr 2023): esma.europa.eu
European Commission, Commission proposes improvements to the SFDR (Nov 2025): finance.ec.europa.eu
Arthur Cox, SFDR 2.0 update (16 Oct 2026): arthurcox.com
CNIL, Blockchain: solutions for a responsible use of the blockchain in the context of personal data (Sep 2018): cnil.fr
European Data Protection Board, Guidelines 02/2025 on processing of personal data through blockchain technologies, version 2.0 (Jul 2026): edpb.europa.eu
Frequently asked questions
Why is impact measurement in affordable housing difficult?
Because the data that impact frameworks depend on sits in each operator's own systems, is reported on the operator's schedule and definitions, and is rarely verified independently. In a GIIN survey of 278 impact investors published in January 2020, 92 percent described collecting quality data as a significant or moderate challenge, and 89 percent said the same of the lack of transparency on impact performance across the market.
Which impact measurement frameworks are most widely used?
IRIS+, the metrics system the Global Impact Investing Network launched in May 2019; the five dimensions of impact (what, who, how much, contribution and risk) developed by the Impact Management Project, which concluded at the end of 2021 and whose resources moved to Impact Frontiers; and the Operating Principles for Impact Management, launched in April 2019 by 60 investors holding more than 350 billion dollars of assets invested for impact.
Does SFDR require funds to audit their impact data?
No. SFDR is a disclosure regulation: an Article 9 product must explain how it pursues a sustainable investment objective, including how it avoids significant harm, but the regulation does not oblige funds to commission audits of operator data. In November 2025 the European Commission proposed replacing the current structure with three voluntary product categories; as of October 2026 that proposal was still in negotiation between the EU institutions.
Can a blockchain verify social impact?
It can verify the financial and operational layer that impact reporting rests on: rent collected, distributions, fees and costs, and whether conditions written into the structure were met, all as one timestamped record that every permitted party reads. It cannot verify human outcomes such as housing stability or wellbeing, and it is only as accurate as the data entered into it.
Can tenant data be stored on a blockchain?
As a rule, no. France's data protection authority, the CNIL, recommended in 2018 that personal data be processed outside the blockchain. The European Data Protection Board's guidelines on blockchain, adopted in final form in July 2026, go further: they state that it is in general not advisable to store personal data on a blockchain, treat hashed or encrypted personal data as still personal data, and recommend keeping it off-chain.