Commercial property price indicators

Venue: European Central Bank, Frankfurt am Main

Commercial property price indicators (CPPIs) are high on the list of policy-makers’ data needs. Recent additional impetus for work in this area has come from the G20, which has identified this dataset as filling a sizeable gap in the statistical landscape; a gap which statisticians need to address, especially in times of financial stress. In particular, an increased use of CPPIs is one of the 20 recommendations listed by the IMF and Financial Stability Board in its Report to the G20 entitled “The Financial Crisis and Information Gaps”. Recommendation 19 calls for the inclusion of “real estate prices (residential and commercial) in the Principal Global Indicators (PGI) website”. The PGI website gives, at a glance, an overview of statistical indicators which policy-makers can use to assess the cyclical situation of the world economy. Recommendation 19 also calls for measurement standards to be developed under the aegis of the Inter-Secretariat Working Group on Price Statistics (IWGPS). Furthermore, the conference has as its origin a proposal made at the 12th meeting of the IWGPS on 5 May 2011 that an international conference on CPPIs be organised to identify and explore issues that will be an integral part of a proposed “Handbook on the measurement of CPPIs”, setting standards for CPPI measurement. Following on from this, Eurostat and the ECB, supported by the International Monetary Fund, the Organisation for Economic Co-operation and Development and the Bank for International Settlements, jointly organised a conference on 10 and 11 May 2012 – see annex – aimed at providing a framework and good practices for the development of such measurement standards.

The conference brought together a large and varied group of participants, both from the user community (policy-makers, financial market users and academics) as well as the producer community (both official statisticians and commercial data providers). It was split into four separate sessions based on the type of stakeholder involved, with representatives from the academic community contributing to all of the sessions.

Session 1 – The user perspective

Papers in this session explained the user requirements for CPPIs, mainly from the perspective of economic analysis. The key uses identified included their use for macroeconomic and financial stability purposes (at both national and international levels) and their use as data to support investment decisions for financial market participants. As regards statistical quality, data users tended to converge toward some common features, such as harmonised definitions, in particular for property types and market segments, robust concepts and good quality standards. Ideally the data would be based on transaction data but, considering the generally low liquidity of real estate markets, in particular at times of market stress, as well as practical issues related to data collection and sampling given the considerable heterogeneity of commercial properties, valuation-based data are also likely to be of importance. Moreover, a comparison of transaction data and valuation data can provide useful insights, also in terms of the potential for an appraisal bias. The use of valuations requires similar valuation systems to be compared or aggregated across regions. In Europe, however, the differences between these systems can be very pronounced, so that users (and in the next session also producers) unanimously called for an initiative aimed at harmonising valuation systems.

Regarding regional aggregation, while national and international indicators are required for policy purposes, more detailed data is needed when it comes to identifying driving factors, in particular from the supply side, and related market developments. Meeting both requirements in a consistent way may call for the collection of detailed data which can then be aggregated at the national and international levels. In addition, placing property types into asset classes would provide a natural breakdown, given that there is broad heterogeneity across such classes. Furthermore, the lack of harmonised concepts significantly hinders the comparability of derived indicators.

The frequency requirement should be quarterly data, which should be suitably timely. In general, the market participants would prefer more granularity in the amount of data made available, while the policy-making community emphasised timeliness with fewer breakdowns. Furthermore, given the choice between timeliness and reliability, several participants stressed the importance of having data available with a short time lag, particularly at the aggregate level.

There was recognition that the needs of national accounts statisticians had also to be explored and reconciled with the needs outlined above.

Session 2 – The producer perspective

Commercial data providers and one central bank explained in this session how they collect their data and for which purposes. Prime data sources differ across countries and providers; they include funds, banks, government agencies, surveys and stock markets. The main user community to date has been the investment community, and consequently most of the indicators currently available in the market tend to be aimed at portfolio benchmarking.

Although some data providers concentrate on one segment of the commercial property market, such as offices, there was agreement that, ideally, all segments should be covered, including, for example, warehouses and retail, industrial and services buildings, but potentially with different prioritisation. Residential property that is commercially managed should probably be included in an indicator for financial stability but is less relevant in a macroeconomic context, where prices are already available from residential property price indices. In the same vein, although some data providers concentrate on valuation-based data for the “prime” locations of the market, there was agreement that the ultimate aim is information on all locations. One shortcoming raised was the lack of agreed definitions for differing quality locations. Furthermore, no catalogue for defining the quality of a building exists but would indeed be beneficial in order to obtain more comparable data.

As was the case in the user session there was support for more harmonisation regarding valuation principles for real estate coupled with a call for a quarterly rather than an annual valuation regime.

The central banking community typically takes a multi-input approach in order to construct indicators from several data sources. The aim has been to obtain the best possible indication of the development of commercial property prices over time at the national level, but the approaches and data sources have not been fully comparable. Nonetheless, a prevalent methodological approach has emerged which exploits any available transaction-based data sources and supplements them with other valuation and/or stock market-based indicators. Based on this line of reasoning and national input, the ECB is aiming to compile an experimental set of CPPIs for the euro area – and possibly the EU – by the end of 2012.

Private data providers use their data to compile a variety of indicators reflecting both the properties of the data collected by them and their interest in the dynamics of transactions, yields and capital value. Additional indicators compiled cover rents and occupancy or vacancy rates.

There was a general call from producers and users for more transparency in the concepts and methodologies underlying the indicators compiled. Providing metadata is seen as a key requirement in order to give users an insight into what the indicators can and cannot measure and whether different developments of indicators over time in different markets reflect different economic dynamics or statistical artefacts.

Session 3 – Official statistics providers

From the perspective of a conceptual framework, the application of the methods of measuring changes in prices over time, as established in official price statistics, to the case of commercial property prices identifies the key issues with which data producers are confronted when creating new indices or improving existing indicators. While meeting these standards in every dimension will be challenging, it helps to decide on the priority areas of conceptual work and to identify which metadata are of particular interest for users.

Official statisticians also presented their practical work on CPPIs. In several areas the work is very much in its infancy and is seen to be particularly challenging. Other examples showed that official commercial property price indices had been successfully established in some countries. The official statisticians see the work as being relevant both to develop indicators in their own right and as an input into the national accounting framework. There were several common issues, such as the statistical boundaries (for example, deciding whether residential property solely produced for leasing should be included), the type of data to be used as a headline indicator, classification frameworks, how to deal with time-based movements of housing stock within sectors, and whether differences in characteristics of properties sampled over time can be controlled for using hedonic, repeat sales or other methods. The scarcity and uniqueness of transactions, particularly in economic downturns, were identified as particular problems; there was a call for international guidance in order to move towards harmonised outputs.

A presentation by an academic and private data supplier proposed an eclectic or “cocktail” approach in which several indicators are used in parallel, bringing together public and private market information.

Given the importance of detailed information by property class, by region and by market segment, a regular provision of such granular information, following the standards established in official statistics for the measurement of macroeconomic phenomena, would be a key contribution to providing a useful statistical toolkit for analysing price changes in commercial property markets. Any analysis would benefit from detailed data; however, policy-makers still need a headline rate to refer to in official communications.

Identifying the shortcomings in the “ingredients of the cocktail” provides useful information for the development of combined indices, e.g. transaction-based indices which make use of valuation data as an explanatory variable in a regression framework. Smoothing techniques can be investigated, e.g. in order to adjust for high volatility in transaction data or over-smoothing of valuations. In addition, lead and lag relationships between different kinds of indicators (stock market-based, transaction-based and appraisal-based) seem to deserve further analysis.

Session 4 – Round table and the way forward

The round table was conducted as an open discussion involving the chairs of the previous sessions, representatives from the sponsoring organisations and academia as well as an active participation from the floor. Overall, the conference was seen as successful in bringing together so many key stakeholders for an in-depth discussion, and in committing the international agencies to a swift follow-up. The following conclusions were drawn:

• The information on commercial property prices is relatively widely available, generally from non-official data providers. However, its reliability needs to be further investigated, and better metadata provided. It will be necessary to be pragmatic as to what can be achieved in both the short and longer term and to guide the user community accordingly. With this in mind a quarterly indicator released around 90 days after the end of the quarter might be a suitable development target. A suitable goal might be to produce a reliable indicator to display trend movements and turning points in commercial property prices.
• In terms of the compilation of indicators, statistical work is likely to have to be innovative since the indicator may not be fully based on transaction-price data. A partnership between official and commercial data providers in this field is seen as mutually beneficial. With this in mind, some of the commercial data providers expressed a willingness to allow the official statisticians access to their detailed datasets. Reconciling the needs of national accounts and commercial property price indices as economic indicators was identified as being potentially challenging.
• There was a call for the definition of harmonised and comparable standards.
• It is likely that a multi-source approach will be required for the development of a headline indicator. Transaction-based data are likely to be the best theoretical approach, but cannot be the sole data source as in times of stress market liquidity tends to be too low. Valuation data, perhaps accompanied by further indicators, are therefore likely to be key inputs. The methodological implications of the multi-source approach will also have to be assessed. Examples of supplementary information include data such as vacancy rates, financing information and rents.
• Supporting metadata should also be produced to enable users to understand the headline data.

Given the above conclusions from the conference, Eurostat, as the lead agency under the aegis of the IWGPS, has already initiated a process aimed at putting together a handbook for compilers of commercial property price indices. Although it still needs to fully cover theoretical aspects, it is expected to serve as a pragmatic, operational “cookbook”. The authors of the handbook will need to consult with the commercial data providers to ensure that the latter’s request for a harmonisation process is covered in a mutually beneficial way. The handbook is expected to become available by end-2014.

Commercial Property Price Indicators (CPPIs) are a new and challenging statistical area for which there is increasing demand. This is expected to lead to important developments in the coming years. These indicators are of high interest for both public institutions and private investors who wish to carry out economic and financial analysis. The lack of data in this area has been highlighted in particular during the recent economic and financial crisis. However, international standards have not yet been developed in this area. Indeed, the concept of commercial property is not yet defined in official statistics and the links with related areas of official statistics, such as price statistics and national accounts, not yet explored. Hence, the future practical and methodological developments for CPPI will benefit from a coordinated approach involving national statistical institutes and/or national central banks as well as non-traditional statistical actors.