New study shows that 10 SCPIs are enough to capture 88% of the benefits of diversification

Written on 28 Jun 2021.


printer-friendly version

Real estate has become an essential part of institutional investment portfolios and is now established as a source of diversification and added value in the context of multi-class portfolio construction. A variety of real estate investment vehicles have been designed over time to match investors’ needs and provide solutions to the challenges that are typical of direct investing. The use of non-listed real estate collective investment schemes has been widely explored and analysed in academic and industry research but there is no published research focusing on French non-listed real estate vehicles to the notable exception of Schoeffler (2020)[1].

The regulated French investment vehicle known as Société Civile de Placement Immobilier (SCPI) is of particular interest and more specifically the commercial SCPI market, which experienced double-digit growth over the last 10 years, with AUM almost quadrupling to reach EUR 67bn by the end of 2020.

In a new publication “Benefits of Open Architecture and Multi-Management in Real Estate Markets— Evidence from French Nonlisted Investment Trusts”, EDHEC-Risk Institute, in partnership with Swiss Life Asset Managers France, assess whether modern investment management techniques such as fund selection and portfolio allocation can be applied to the SCPI universe[2] and create value for an investor wishing to be exposed to French non-listed commercial real estate.

 

Key findings of the report[3] include the following:

 

SCPI RISK AND PERFORMANCE ANALYSIS

  • Analysis of the time-series performance of the EDHEC IEIF Commercial Property (France) Index shows over 75% of the index’s total return throughout the period is explained by dividend distribution, which is consistent with the purpose of SCPIs as rental income-generating vehicles.
  • Despite being the largest contributor to total return, the dividend yield plays little role in differentiating across SCPIs’ performances; the marginal contributions of price return and dividend yield to the dispersion in performance across SCPIs[4] are respectively equal to 89% and 11%;
  • The main driver of outperformance for a given SCPI is therefore the average price return, which varies as much as the total return (going from -0.6% to 7.6% on average annually) across the universe of SCPIs;
  • Price return plays a similarly important role in explaining the large discrepancies in risk across SCPIs;
  • The large dispersion observed in performance and in risk naturally supports the idea that investors would benefit from the implementation of selection and allocation processes.

 

PORTFOLIO DIVERSIFICATION AND BENEFITS OF ALLOCATION 

  • As expected, increasing the number of SCPIs in a portfolio does on average reduce its total return volatility;
  • The equally-weighted SCPI portfolio[5] indeed carries significantly less risk than the “average SCPI”; it appears that 88% of the full risk-reduction is obtained after only 10 SCPIs: eliminating over 80% of the population from the portfolio does not, on average, substantially reduce the benefits of portfolio allocation for investors;
  • Diversification comes at a cost given the frictions prevailing in the SCPI market, and investors may need to consider the tangible benefits of allocation as part of a trade-off.

 

RELEVANT CANDIDATE ATTRIBUTES FOR SCPI SELECTION 

The analysis covers eight SCPI attributes, and the authors identify three specific candidates that could help explain the cross-sectional differences in risk and return:

  • Fund size attribute (ranking SCPIs by AUM) : small SCPIs have historically outperformed their larger peers;
  • Volatility attribute (ranking SCPIs by time-series volatility): low-volatility SCPIs have produced strong in-sample risk-adjusted returns compared to their higher-volatility peers;
  • Past Performance attribute: past losers have historically tended to underperform in subsequent periods.

 

Although further research will be needed to formally validate the three candidates (including rigorous out-of-sample testing and analysis of SCPIs’ financial statements), current results indicate investor welfare could be enhanced via suitable SCPI selection decisions.

 

Overall, results suggest that value can be added by selection and allocation decisions, which could form the basis of a welfare-enhancing open architecture multi-management approach to investment in SCPIs.

 

Commenting on this research, Lionel Martellini, Director of EDHEC-Risk Institute, said, “Modern asset management tools can easily be applied to the real estate asset class. In particular, a multi-management approach on the SCPI market based on the right selection and allocation processes is likely to significantly improve the risk-return profile to the benefit of investors.”

 

Frédéric Bôl, Chief Executive Officer, Swiss Life Asset Managers France, underlined: “The collaboration between our respective research teams enabled us to adapt the tools developed in the context of financial asset management to the specific universe of non-listed real estate investment funds, and SCPIs in particular. The results of this research are directly operational with a view to improving the SCPI investment process both for individual savers and institutional investors.” 

 

Commenting on this research, Béatrice Guedj, Head Research & Innovation at Swiss Life Asset Managers France, said, “Given the increasing appetite for SCPIs vehicles, a building block for a long term investment, both households and institutional investors should remember: firstly, risk-adjusted performance is driven by price return not by yield; secondly, selection on size, volatility and track record are relevant for an optimal allocation process. Thirdly, material diversification, as chased by both savers and investors, would be achieve through a portfolio of 10 SCPIs. The challenge is now to find the best route to efficiently implement such an open architecture multi-management solution for the benefits of all.”

 

Commenting on the results of the Paper, Shahyar Safaee, Research Engineer, EDHEC-Risk Institute: “The analysis of the risk-return profile of SCPIs reveals a diverse and heterogeneous universe of real estate investment funds where long-term performance dispersion, contrary to what one might think, is primarily due to differences in price returns rather than dividend yields. Such significant performance dispersion should encourage investors to implement an allocation process. In particular, we observe a decrease in the average risk of SCPI portfolios as their number of constituents increases, and the fact that as few as 10 SCPIs are sufficient to achieve significant diversification raises the question of selection. To this end, investors could look at observable and differentiating SCPI attributes such as size or past performance to build a value-adding selection process.”

 

A copy of the publication can be downloaded via the following link:

EDHEC-Risk Institute Publication: Benefits of Open Architecture and Multi-Management in Real Estate Markets—Evidence from French Nonlisted Investment Trusts

 

You can access the editorial – A Case Study in Real Asset Investing: the French Non-Listed Real Estate Fund Market – published in the April edition of EDHEC-Risk Institute quarterly newsletter.

 

This research was supported by Swiss Life Asset Managers France as part of EDHEC-Risk Institute’s “Real Estate in Modern Investment Solutions” research chair.

 

 

[1] Schoeffler, P. (2020). Liquidity of real estate funds available to the general public in France. Association française des Sociétés de Placement Immobilier (ASPIM).

[2] Our analysis focuses specifically on SCPIs invested in commercial real estate. They have the longest track record and represent 94% of the market in terms of assets under management (AUM).

[3] This study focuses on a representative set of 53 SCPI vehicles between 2003 and 2019. Data was kindly provided by the Institut de l’Epargne Immobilière et Foncière (IEIF).

[4] The average annual total returns over the sample period vary from 5.2% to 13.4%, a significant discrepancy over a 16-year horizon.

[5] With 53 constituents.