Suggested applications of tokenisation in real estate finance

Of the three concepts analysed in this book (factorisation, digitalisation, and tokenisation), tokenisation is the least innovative conceptually even though it relies on a cutting-edge technological apparatus. Nonetheless, provided the right applications, tokens might play an important role in the future of real estate finance by offering a tool that financial engineers can apply to real assets. Baum (2020) underscores that “if real estate tokenisation were to become popular, several innovations become conceivable, including structured finance, hybrid real estate tokens and digital future exchanges”. The following section introduces two such innovations: first, index tokenisation which is the application of blockchain-based tokenisation to indices of private commercial real estate used as underlying to real estate derivatives, and second, data tokens which are utility tokens on data collected in smart buildings. The objective in introducing these innovations is to explore how tokens can be used to address two of the real estate sector’s needs:

• - the long-standing need for better risk management instruments given indexbased real estate derivatives’ shortcomings mentioned in Chapter 1,
• - and the emerging need to optimally allocate and safely monetise real estate resources in smart buildings, such as data, as explained in Chapter 2.

Index tokenisation

Establishing a successful market for index-based real estate derivatives has been hindered by issues that real estate researchers have found challenging to overcome, including:

- The lack of granularity in the spot commercial real estate market which “may restrain the trading of commercial real estate derivatives [...]” (Tunaru and Fabozzi, 2017),

• - The lack of replicability of the underlying index inhibiting “banks from launching derivative product and simply replicating it by buying the properties contained in the underlying index” (Syz, 2008),
• - The numerous frictions in the real estate derivatives market “because the index and its components cannot be traded continuously and instantly at the prevailing spot price without transaction costs” (Syz, 2008),
• - The impossibility to implement dynamic hedging strategies using real estate derivatives since “it is hardly possible to trade the constituents of a property index [which] are usually indivisible and not traded in small units” (Syz, 2008).

Considering these shortcomings, Syz (2008) explains that “if a property index can be decomposed into factors, some of which are traded assets, then the index can at least in part be mimicked”. This is where, in addition to factorisation analysed in chapter 1, tokenisation can add value to real estate derivatives, by contributing to establish a more liquid market for real estate derivatives. Several alternative formats can be envisioned. The following paragraph reviews three alternatives.

Real estate derivatives on single asset tokens: the first alternative is to design real estate derivatives on single asset tokens. However, given the expected narrowness of single asset tokens’ spot markets, it is highly unlikely that the regulatory authorities (e.g. Securities Exchange Commission and Commodity Futures Trading Commission under the 1982 Shad-Johnson accord in the USA) will easily approve such instruments. As pointed out by Shiller (1993), the risk of manipulation in the cash market is high and challenging to curtail for commercial real estate. Notwithstanding their intrinsic challenges, real estate derivatives on single asset tokens can make sense as part of combinative hedges alongside other tradable assets and/or factors.

Real estate derivatives on index of tokens: Won (2019) mentions the creation of indices of tokens as a possible future application of tokenisation. Indices of tokens, i.e. indices made up of single asset tokens, could be used as underlying to real estate derivatives in lieu of existing indices of institutionally owned commercial real estate (e.g. MSCI-1PD indices). In principle, this alternative could solve the above-mentioned issues of granularity, replicability, and frictions of property indices. In practice, it supposes that large and liquid markets for single asset tokens operate. Even if it were the case, tokenised properties available for constructing indices of tokens may not be of institutional grade and comparable to properties included in valuation-based indices of commercial real estate, therefore limiting indices of tokens’ representativeness and contribution to investors’ hedging strategies. Interestingly, indices of tokens can open the door to index customisation. With n tokens, n! customised valuation-based indices can be created. Issues of governance, especially as far as periodic valuations of the underlying properties are concerned, have to be ironed out before such a model can be implemented. Customised indices of tokens may best serve as constituents in combinative hedges as mentioned previously for single asset tokens.

Real estate derivatives on tokenised indices: both real estate token-based derivatives models presented before suppose a functioning market for single asset tokens. This is clearly a limitation. The third alternative presented hereafter does not rely on existing single asset tokens. Instead, it relies on a process defined as index tokenisation. While indices of tokens pool single asset tokens into an index, index tokenisation creates security tokens from off-chain properties included in an index (e.g. NCRE1F Property Index). These security tokens known as index tokens are designed to mimic off-chain properties’ returns. Returns on private commercial real estate indices are thus fractionalised into a series of index tokens, each token replicating the individual return(s) of one actual off-chain property in the index. Hence, index tokenisation does not involve fractionalisation of off-chain properties nor the on-chain transfer of off-chain assets. Only index returns are fractionalised into property-level returns through the creation of on-chain derivative securities, i.e. index tokens which are blockchain-native and whose values derive from off-chain returns. Figure 3.2 illustrates the concept of index tokenisation.

Index tokenisation can be implemented on a perpetual basis (one token per off-chain property for all periods) or on a periodic basis (one token per period for each off-chain property). Assuming that several blockchains can be opened in parallel,

Notes: (i) Property returns (e.g., total returns, market values) are periodically computed by the index provider according to the index' methodology (e.g., Young, Fisher and D'Alessandro, 2017 for the NCREIF Property Index). These returns are then weighted (e.g., Wi for property i might be an equal weight or a valuation-based weight) and released as an aggregate periodic return in % for the index, (ii) Index tokens are security tokens based on fractionalised index returns rather than fractionalised properties. For properties in the index of private commercial real estate, there are n! possible indices of index tokens.

Figure 3.2 Index tokenisation for private commercial real estate indices.

Tokenisation of commercial real estate 151 the latter format enables more flexibility and trading opportunities, by opening up a range of periodic maturities (e.g. quarterly returns tokens) to market participants. The index tokens market involves blockchains for easy order matching and position clearing, as well as smart contracts for transacting. For starters, private blockchains are made available to accredited investors only. Both long and short positions can be allowed provided transactions find counterparties on the chain. The risk of insufficient liquidity obviously looms large on such a market, which may eventually justify the opening of the market to both hedgers and speculators on public blockchains.¹⁴

Utility tokens for smart real estate: data tokens

Chapter 2 underscores the emergence of new resources in smart buildings in smart cities. Among these resources, data are expected to play a paramount role in creating value for property owners. According to the model of digital rights, ownership of data collected in a smart building is based on Digital Usage Rights. Data collected in tenanted spaces belong to tenants whereas data collected in common areas and non-tenanted spaces belong to property owners.

Property owners who wish to monetise their share of collected data can ‘rent out’ some data to a third party, e.g. one of their tenants. One way to do so is to set up a private blockchain allowing property owners to sell utility tokens on data or data tokens. Data tokens are attached to a specific use of data during a limited period of time, e.g. selling to a tenant the right to use data collected in a shopping mail’s common areas during a week or at certain times of the day. The same model of data tokens can apply for tenants looking for ways to monetise data collected within their rented spaces. Any innovation involving data collected in space is sensitive and obviously has to abide by data protection and privacy laws.

In sum, utility tokens can complement the concept of digital rights and help establish a transparent and frictionless rental market for smart real estate’s resources, thereby allowing for a finer allocation of value created in smart buildings.¹^

Notes

1 Roche (1995) defines Single Property Ownership Trust (SPOT) as

a securitisation arrangement which would have provided the investor with a specific share of a single property’s income flow and capital value via a trust mechanism. Strictly an investor would have held units in the trust rather than securities in a company.

AProperty Income Certificate, or PINC, is “a tradable equity security in a tax transparent single property company which entitles the holder to a direct share of the property’s income flow and capital value. A PINC company could be geared, [but] never traded”.

• 2 Tire notion of democratising access to property markets owing to tokenisation is often associated with a range of sociological goals defining a lifestyle where property ownership does not matter as much as before, e.g. shared ownership of properties. Ernst and Young (2020) talks about inclusive finance and highlights that fractioning assetsallows “multiple people [to] buy together an asset and use it, which is key to a society where usage is more and more supplementing ownership”. These objectives are commendable but hardly new. In real estate, the concept of having partial use commensurate with partial ownership is called timeshare. More generally, what people actually pay for when they buy or rent properties is time to appropriate space as pointed by Lefebvre (1974) in his analysis of real estate tenures: eternity in case of freehold and shorter periods (leaseholds) with possibly idiosyncratic periodicity (timeshare arrangements) for other tenures . Tokens are simply encapsulating time in a digital format.
• 3 In that respect, asset classes such as fine art, wines, and other collectibles whose off-chain lives are a lot less eventful than buildings’ (no cash flows, no depreciation and limited interactions, if any, with their environments under optimal conservation conditions) would be potentially more suitable for tokenisation than commercial real estate assets. Baum (2020) notes that “artwork does not suffer depreciation or obsolescence, or need refurbishment, or deliver income, making it somewhat less complex as a fractionalised asset than real estate”. Despite tokenisation’s promise of a fully decentralised market, the tokenisation of fine art still requires custodianship by a central authority or trusted third party. Custodianship is an essential characteristic of real asset tokenisation as highlighted in OECD (2020).
• 4 A regulated exchange called 1PSX (The Property Stock Exchange) has been set up in London in 2019 in order to trade single real estate asset companies, albeit not security tokens nor tokenised securities, insofar as IPSX does not use blockchain or DLTs.
• 5 The application of blockchains in commercial real estate should contribute to improve the property search process (Deloitte, 2017), by aggregating on-chain fragmented listings data from multiple listings services.
• 6 In theory, provided their markets are liquid and efficient enough, security tokens are expected to generate lower expected returns for their holders than private commercial real estate since they hold less risk (ceteris paribus).
• 7 A privately owned asset can be off-chain (physical) or on-chain.
• 8 The view that blockchains through consensus can bring out a dimension larger than individual agents’ contributions is in line with Pierre Lévy (1994) who describes collective intelligence powered by digital technologies (e.g. the internet) as a trigger to a “real time democracy”.
• 9 In a seminal text, Habermas (1973) explains the concept of consensus-based truth as follows:
• 1 may ascribe a predicate to an object if and only if every other individual who could enter into discussion with me would ascribe the same predicate to the same object. In order to distinguish true from false statements, I refer to the judgement of others- in fact the judgement of all others with whom I could ever undertake a discussion [...] The condition for the truth of statements is the potential agreement of all others [...] Truth means the promise of achieving a rational consensus.
• 10 Specifically designed governance principles should regulate tokens at every stage of their investment cycle, from offering to redemption. For instance, regulators should define the rules governing investors’ exit strategies, e.g. “en-bloc” sale of all tokens to a third party based on a quorum system.
• 11 In the real estate tokenisation discourse, the mitigation of asymmetric information and the démocratisation of commercial real estate markets are usually closely intertwined. The lack of transparency supposedly benefits large investors who appropriate most of the value created in commercial real estate, whilst small investors are left out. This argument seems ideological rather than economic unless blockchains are actually meant to trigger a shift out of the Western capitalist model to a new type of model which remains to be properly defined.
• 12 In terms of risks, single asset tokens are potentially riskier than REITs due to their narrower focus on one property only. In contrast to securitisation (e.g. MBS), tokenisation does not strive to restructure assets’ cash flows. Tokens’ returns directly reflect, without

Tokenisation of commercial real estate 153 filter, the underlying property’s life cycle. As the building ages, secondary tokens offerings might be required to finance capital expenditures, thus diluting existing tokens holders’ interests. Due to the importance of property management in the real world, security tokens may end up being externally managed vehicles, an eventuality which underscores the key role that governance must play in real estate tokenisation.

• 13 In practice, a strict application of the Arrow-Debreu model is unrealistic, a point mentioned in the introduction of this book (after Shiller, 2004). However, a tokens market which involves no financial engineering does not suffice to manage risk for property hedgers.
• 14 As a prerequisite condition to index tokenisation, index sponsors should be willing to share some granular, albeit anonymised, data with the public, which may be way more than the level of disclosure agreed upon with their data contributing property owners.
• 15 Depending on use “which can evolve toward or away from a security” (Token Alliance, 2018), utility tokens are sometimes deemed as securities by market authorities. One important characteristic for data tokens to qualify as tokens and not securities would be that there is no secondary market for them: once purchased, they have to be immediately used rather than stored and resold.