A complete, up to date, standardised dataset, identifying all the key investment factors, considerably aids the analytical process, but missing funds and data points, old data, poor data entry and non-standardisation of metrics reduces the value of the dataset to the industry and deters new investors from investing in the market.
The researchers found that there were considerable deficiencies in the quality of the available data and this, at times, limited the amount and complexity of the analysis undertaken.
Funds specified a mix of strategies in the PFV Handbook to achieve their objective: stock selection, asset allocation, active/asset management, development, inflation linked/fixed uplifts leases, covenant strength and lease length (one even listed research!). The research team felt that not all of these strategies can be measured directly with the current data in the PFV Handbook. Indeed, any future revisions to the content should seek to map directly to the stated strategies in order to be capable of subsequently testing their effectiveness.
Five key drivers affect fund returns: cash holdings, leverage, fees, structure and style. The PFV Handbook should provide an authoritative statement on the relative importance of each.
In addition to the quarterly performance and NAV reported already, interest received on cash and interest payments on debt should be added to the PFV Handbook and the influence of cash and leverage on Index and fund returns should be clearly stated. This would complement a complete and consistent statement of the impact of fees (we heartily commend the work of the AREF Fee Working Group to harmonise the reporting of fee levels). It is also unclear as to the robustness of the fee information included in the PFV Handbook. For instance, where funds were in the samples for both 2007 and 2019, the text on fees was identical for the vast majority. This raises the issue of whether the text section of the document is regularly checked for accuracy.
The match between the segmentation utilised in the PFV Handbook and MSCI’s direct property indices is a powerful combination, for subscribers to both services, ensuring consistency between the reporting of the portfolio structure and the past performance characteristics and current pricing of each category. To improve analysis further, the ‘other’ category should be split to reflect the growth in previously very small property types and all measures (e.g. number of properties and income security) should be reported at the segment, not portfolio, or other aggregate level.
Lease types (indexation, ground rents) should be added to the PFV Handbook to distinguish the investment strategies of Long Income funds.
The original rationale for the PFV Handbook was to streamline the process of collecting fund data and therefore saving investors and managers time as well as ensuring the standardisation and quality controlling of the data collected. The fact that it is no longer seen as a pre-requisite, or as the sole means of communicating with investors, suggests that it has not kept pace with investor requirements. Data of this kind and value should never be seen as a static set of questions to be collated for ever, but an evolving source of information required by investors.
There is a fine line between the benefits of collaboration and standardisation, and stifling innovation. The balance seems to have swung too far the other way: it should be the use of data in the investment process that distinguishes one investment house from another, not the volume of data collected. The industry needs to recreate John Atkin’s rigour and enthusiasm for providing a comprehensive set of data on funds for investors. The philosophy underpinning the PFV Handbook and the Index, is that the benefits from collaboration outweigh individual gain from acting alone, have seemingly been so eroded, that a renewed consensus must be found.
Moving the PFV Handbook format, from paper, to pdf and now to a spreadsheet (although the current spreadsheet is unstructured which severely inhibits its useability) is a natural progression as technology has transformed the way we work. This progression can never be complete as our business activities continue to evolve. There are now new ways of working with data, and spreadsheets are rapidly being replaced with more automated tools. The production and dissemination of data should be continuously upgraded in advance of, and in-line with, investor requirements.
There will inevitably be tension between public and confidential datasets and investors’ demand for complete transparency. This tension is compounded by the tendency of managers to cite commercial sensitivity as a reason for the non-disclosure of data, such as current valuations. Transparency is a perennial issue in real estate markets around the globe. Researchers will argue that greater transparency lowers the risk premium and benefits the entire market. Managers will argue that full transparency reduces their ability to generate superior performance. There is probably truth in both arguments and this tension has to be managed. It is ironic that most managers subsequently collaborate with advisory firms and/or pay for data subscription services that allow them to gain access to confidential data – arguably removing any potential competitive advantages in the industry. Private data is increasingly available via different platforms from government (eg HM Land Registry) to private businesses and this will continue, fuelled by technological advances with machine learning and Artificial Intelligence which permit the mechanisation of the collection, storage and analysis of data.