The importance of high quality cross-border portfolio investment statistics
Keynote address José Manuel González-Páramo, Member of the Executive Board of the ECBInternational Conference on the Coordinated Portfolio Investment Survey,Madrid, 1-2 March 2006
With great pleasure I have accepted the invitation by the organisers of this Conference, the International Monetary Fund and the Banco de España, to address you here today. This is because, as a representative of the European Central Bank, I am convinced of the importance of today’s topic for policy-making. So, on behalf of the European Central Bank, I would like to thank the coordinators of the Conference for the invitation, and above all I would like to thank the organisers of the Coordinated Portfolio Investment Survey for their vision and efforts.
This morning, I would like briefly to develop some thoughts on the importance of statistics for policy- making, that is to say from the viewpoint of a user of statistics, and in particular on the importance of the portfolio investment statistics, which will be the core subject of this conference.
Importance of statistics for policy-making
The Governing Council of the European Central Bank has at its disposal a wide range of statistics for monetary policy-making. It is important to stress that the statistics I am referring to are neither a single model nor a limited set of key indicators. On the contrary, these statistics comprise a comprehensive, integrated and consistent set of data for the euro area. After all, the ECB Governing Council, as any central bank decision making body, or as other policy-makers, operates in an environment of high, and even increasing, uncertainty regarding the functioning of the economy as well as its prevailing state and future development.
In this context, a rich set of timely statistical data is a precondition for sound monetary policy-making. However, at the same time, the information needs to be structured and analysed in a consistent way. At the ECB and within the ESCB enormous efforts have been undertaken since the establishment of the forerunner of the ECB, the European Monetary Institute, to develop such a structured set of statistics for the euro area. As a result, a wide range of consistent, high-quality statistics have been developed and are now produced on a regular basis.
Financial statistics are indeed at the heart of the statistics compiled by the ECB/ESCB and used by the Governing Council. Changes in broad money and their counterparts, evidence of the transmission of monetary policy decisions to the economy, interest rates and exchange rates changes, and transactions with non-residents are key phenomena in the assessment of monetary policy. The analysis covers the monitoring of domestic as well as international markets. The statistical information is typically produced and analysed at the macro level, in the form of aggregated statistics describing the euro area as a single economic entity. Outstanding examples of macroeconomic financial statistics are integrated euro area financial and non-financial accounts, a high priority in the ECB’s current development of statistics, and the balance of payments and international investment position statistics. The ultimate goal of these macroeconomic statistics is to cover both stocks and flows of all financial assets and liabilities of the various sectors of an economy and to reveal at an aggregate level who is financing whom.
Negotiable instruments in general and securities in particular are a core segment of the financial markets, if only because transactions in securities account for a large share of global financial transactions. Indeed, cross-border transactions in securities between euro area residents and residents of countries outside the euro area have significantly affected developments in M3 in recent years. Furthermore, outstanding positions in securities represent a major share of the total external financial exposure, as reflected in the international investment position of the euro area vis-à-vis the rest of the world.
To illustrate the importance on data on securities, let me use the most recent statistics published by the ECB. At the end of 2005, debt securities issued by euro area residents amounted to10 trillion euro, a quarter of which was held by non-residents. Non-residents also held 4 trillion euro of equities (in direct and portfolio investment) and euro area investors held foreign securities worth 5 trillion euro. In addition cross-border income streams (associated with direct investment equity and portfolio investment) amounted to 16% of goods exports and 21% of goods imports in 2005. Moreover, securities represent about 20% of the monetary financial institutions’ (MFI) consolidated balance sheet in the euro area, thus making an increasingly important component of broad money.
Given these orders of magnitude, it should not come as a surprise that security statistics play an important role at the ECB. For example, the ECB publishes monthly securities issues statistics, and transactions and positions in securities are reflected in the consolidated balance sheet statistics of the monetary financial institutions, in monetary aggregates, in the euro area financial accounts and in balance of payments and international investment position statistics.
Statistics on portfolio investment
In view of the increasing globalisation and financial integration, the ECB has also strongly supported the IMF’s initiative to organise the first Coordinated Portfolio Investment Survey, the CPIS, and to hold an annual survey since 2001. The CPIS now provides a reasonably reliable geographical breakdown of most securities held as portfolio investments and is a unique source for actual cross-border creditors’ holdings. However, the CPIS survey is still only annual and the results only become available with a delay of more than a year, which seriously limits its usefulness for policy analysis in real time.
More generally, it is fair to say that currently our assessment of the quality of portfolio investment statistics is not entirely positive. This conclusion has been substantiated in the context of our regular ECB reports on the quality of euro area balance of payments statistics, most recently in the edition of this report that was published two weeks ago. This applies not only to the euro area balance of payments, but worldwide. In this context, I should also refer to the sizeable global discrepancies in flows and stocks which are revealed in the annual balance of payments report of the IMF.
Against the back top of this assessment, the ECB and the ESCB have gone a step further to improve portfolio investment data, which may also benefit the CPIS. This concerns the collection and compilation of financial statistics in general, and portfolio statistics in particular, on the basis of so-called “security-by-security” reporting. In a nutshell, security-by-security reporting means that (cross-border) transactions and/or holdings of individual securities are reported. For each security, the residency of the foreign counterpart, the sector of the resident counterpart and a number of other features must then be known in order to compile the relevant statistics.
A complication in this regard is that whereas such information is commonly available from commercial data providers, it is typically neither complete nor consistent across providers. For that reason, the ECB has built a comprehensive reference database, the Centralised Securities Database, which covers all attributes attached to a wide range of securities, and which ensures that for each security a single, complete and reliable representation exists, the so-called ‘golden copy’. This golden copy is constructed on the basis of the raw material provided by commercial data providers, through advanced algorithmic procedures that require minimum human intervention. The system is supported by an ESCB-wide data quality management. The CSDB has been operational since May 2005.
Security-by-security reporting will substantially reduce the response burden for reporting agents. Detailed information on the individual securities held or traded is readily available in the databases of major market players that are reporting agents for these statistics and they can thus automate the reporting of their raw data (i.e. securities identifiers, amounts traded / held, etc.). By contrast, the reporting of aggregate data implies much higher costs to respondents as such aggregations are typically not used for internal purposes and must thus be done for statistical purposes only. An additional advantage of security-by-security reporting is that any future changes in the statistical classification or valuation criteria will not anymore have a spill-over effect on the data to be transmitted by the reporting agents and thus the concomitant investment costs are saved.
The Centralised Securities Database is meant to contain the best possible representation of all individual securities that are issued or potentially held by euro area resident entities. We would, nevertheless, welcome further international cooperation to enhance the quality and coverage of securities issued outside the euro area.
Importance of international cooperation in statistics
Let me stress a final point, that is, the key role played by international cooperation in the improvement of statistics. Evidently, the Coordinated Portfolio Investment Survey is a good example, since this is indeed the result of intensive and fruitful cooperation between the IMF and all participating countries. Of course, a fortiori the importance of cooperation applies to the euro area statistics, which are the result of an intensive process of methodological harmonisation, in line with the international statistical standards, and are being compiled through a streamlined cooperation of the Central Banks of all participating Member States and the ECB.
It may be noted that the application of international statistical standards is a necessary but not a sufficient condition to arrive at consistent global statistics. For example, countries will only be able to share a common analysis of their bilateral economic and financial relationships if their mutual external statistics mirror each other. Unfortunately, however, a recent study by the ECB comparing the mirror data for direct investment between the euro area, the UK, the US and Japan has revealed that sizeable asymmetries between the statistics of these areas still exist. Even within the euro area, the bilateral balance of payments statistics suffer from such quality problems, although these cancel out to a significant extent at the euro area aggregate level. These discrepancies will be eliminated in the forthcoming euro area sector accounts, which are the euro area pendant of the national accounts for institutional sectors. As a consequence, though, the euro area accounts will no longer be exactly equal to the sum of the national accounts of the participating Member States. Instead, they will reflect the best statistical representation of the euro area economy
Let me now conclude. In today’s globalised economy, integrated sets of timely and reliable statistics play a key role in policy-making as well as in investors’ decisions. Such statistics and especially statistics on cross-border transactions and positions require an intensive international cooperation. A good example of this fruitful cooperation is the efficient and effective coordination between the statistical departments of the ECB and of the national central banks. In this regard, the ECB’s Statistics Committee has been very conducive to the development of a rather comprehensive set of euro area statistics that is regularly compiled and disseminated by the ECB. The CPIS is certainly one of best examples of global statistical cooperation, under the umbrella of the IMF. I can assure you that the ECB will wholeheartedly continue to participate in this endeavour and would even be in favour of investigating the possibilities to further enhance its timeliness, frequency and scope.
I wish you a very successful conference.
 The most recent international investment position statistics refer to the third quarter of 2005.
 See ECB Monthly Bulletin, Euro area statistics, Tables 4.1, 4.2, 7.1.3 and 7.4.