Management of foreign exchange reserves

Finally there are the portfolio managers, who will be responsible for carrying out the instructions of their line management and providing the day-to-day management of the asset portfolios.  This is done by active portfolio management, ie direct investment operations with the market within agreed guidelines of flexibility and authority.  It is a feature of this devolved delegated approach that it is typically only the portfolio manager level that has direct contact with market counterparties. For this hierarchical approach to be effective, it is important that each level is clear what it is responsible for, and also that senior levels of management avoid seeking to reclaim for themselves decisions that have been delegated to a lower level.  This is best achieved through a formal benchmark process.

The process of delegation and control should be completed by a regular review and reporting schedule.  This will enable senior management to retain the overall control of the reserves management process.  If the structure is designed properly, it will also identify which part of the overall return stems from which decision, thus enabling all levels to see directly the results of their own decisions.  This direct and highly visible connection between decisions taken and profits earned is both essential feedback in analysing performance, and also an excellent motivator for junior reserves management staff.

If, on the other hand, senior management retains direct control of investment decisions, junior staff will act merely as order-processors, with little incentive to add value to the reserves management operation.  As well as being likely to result in missed profit opportunities, this is demotivating for the junior staff involved and fails to develop the next generation of senior managers.

Description of the benchmark process

The most commonly used method by which decisions from one level are passed down to the level below, and by which results of decisions are calculated and passed back up the management chain, is the Benchmark process. A benchmark is a notional or imaginary portfolio constructed to provide a yardstick or baseline against which the return on an actively-managed portfolio can be measured.  In most cases the composition of the benchmark will be a senior management decision, and it will be set up to conform to a given size, credit quality and average maturity.  Benchmarking provides a base or neutral position from which a manager can adopt his own position according to his judgment and market views, and acts as a comparison point against which to record the results of those decisions[7].

There are three fundamental properties of a benchmark-based decision process. Firstly, the decisions of one layer of management form the benchmark for the next layer down.  For example, the senior management of the central bank may agree a core strategy of keeping 50% of the reserves in US $ and 50% in euro.  This then becomes the base or core position around which the line management of the reserves management team operates.  They in turn decide such matters as whether to hold a small long or short in $ versus euro against the core strategy.  Their decisions on these points will then form the base positions from which the active portfolio management is run.

Secondly, it must be possible physically to match one’s benchmark.  This is true whatever level of management is being considered.  A manager is only responsible for the amount by which his own position deviates from his benchmark.  It follows that if he has no firm views on an element of the portfolio, he will wish not to hold a position there;  and this is achieved by matching his benchmark.  For example, the line management may have no views on, say, the future direction of currencies. With no views, there is no point in holding a position (as to do so is to run a needless risk).  The line management will therefore match the senior management’s currency position exactly. Finally, the results attributed to a manager must relate only to his position relative to his benchmark.  A manager, at whatever level, should not be responsible for a decision that was not his to make[8,19]. 

Alternatives to benchmarking

The complexity of the benchmark process, and the need to construct a bespoke benchmark for one’s own particular circumstances, has led to a number of less complicated alternatives being considered for use in official reserves management. The most common three are indexing, comparison to external managers and targeting a fixed rate. Indexing is the most common alternative. There is a wide range of public indices available for a central bank to match its portfolio to, with many of the investment houses producing them and most available on the wire services or electronically.  This is essential if detailed analysis is to be done of positions relative to the index.  The main advantages of indices are[8, 27]:

−  transparency – the index is publicly available and there is no doubt about its properties such as its duration, composition or value;

−  external measurement – the index returns are calculated by the index provider and cannot be disputed;

−  simplicity of use – using an index avoids complicated benchmark calculations.

Against this there are two major disadvantages with indices.  First, they cannot reflect a central bank’s individual circumstances.  Indices are of necessity general, and may contain instruments the central bank does not want to (or is not permitted to) buy.  Or they may not match the desired currency or duration position that the central bank wishes to adopt.  Although indices can be tailored to overcome these difficulties, much of the advantages of simplicity and transparency are lost in doing so. Second, indices can be too comprehensive and lead to too much trading.  Some indices contain hundreds if not thousands of securities and are rebalanced with great frequency.  A portfolio manager trying to match such an index is faced with the choice of multiple holdings and trades, which may not be efficient, or holding a subset of the index, which introduces the risk of different performance from the index (known as “tracking error”).  Often the only solution to this dilemma is a compromise between the two which is not entirely satisfactory on either count. An alternative approach which many central banks consider is to compare the internally managed portfolios to a portfolio given to an external manager.  The main advantage of this method is that it ensures that the comparator for the internal portfolios is realistic – it is itself an actively managed portfolio and faces all the same opportunities in the market to add value.  But the approach has a number of disadvantages[4,24]:

−  the comparison will only be fair if the external manager has exactly the same opportunities (investment instruments etc) and faces exactly the same constraints and limits as the central bank itself;

−  the comparison will reflect the performance of the external manager as much as the central bank’s;

 −  any analysis of the central bank’s own performance will therefore need considerable amounts of information on the performance of the external manager to understand the reasons for the relative performance;

−  the process does not help establish what the external manager’s benchmark should be;

−  the process cannot be used by central banks which do not want to (or are not allowed to) entrust money to external managers.

A final method of assessing the performance of the reserves management operation is to compare it to an absolute target.  For example, the target could be for a minimum return of say 5%.  This has the advantage of simplicity.  But it has many disadvantages.  The main one is that it takes no account of general market conditions – if market rates are high, or if the trend of prices is favourable, the target is too easy to beat, and so does not set a demanding test of performance, while if rates are low, the target is very difficult, and can tempt managers to increase risk too much to match it.  Moreover, such a target gives no opportunity for senior management to express a view about the overall direction they want the reserves management operation to pursue.

Because of these major disadvantages, few central banks will rely exclusively on absolute targeting for measuring their reserves management.  However, more for psychological reasons than for any more financial ones, it is less uncommon for a central bank to include an absolute target as one of the objectives for their reserves management.  For example, a common one is that the overall absolute return on the reserves should not be negative over the year.  This can have merit if it prevents a public loss of confidence in the reserves management process which a net negative result might produce.

 

2.2. The importance of reporting

Given the great degree of delegation in reserves management, the importance of maintaining overall responsibility and control as the counterpart to this delegation has already been observed.  As well as a formal structure of decision-making and a formal monitoring system to ensure that limits are adhered to, this requires a comprehensive reporting system, through which senior management can observe the consequences of the investment decisions their portfolio managers have undertaken. However, portfolio performance reporting is not simply the method by which senior management see how much return the portfolio managers have earned, important though this is both in monetary terms and for more general staff appraisal and management purposes.  It is also the way in which senior management can assess their own decisions. It provides the base data for more public reporting and accountability, for example to parliament.  And through published data and the statistics on reserves holdings given to the IMF it adds to the data available to those pursuing international financial stability and can act as an early warning of financial strain on a country’s foreign exchange position[9,12].

  Internal reporting.An internal reporting system should be regular, frequent, and timely.  Reports should be regular so that there is no possibility of awkward or unpleasant news being covered up.  They should be frequent so that management can maintain close control and stop a situation getting out of hand before it goes too far.  And they should be timely (ie, reporting should be as soon after the period being covered as possible) to ensure that if there are problems senior management can act before serious damage is done. The content of the internal reports will be largely for each central bank to decide for itself.  But as a minimum, internal reports should cover the external environment, the portfolio manager’s response, and the results of actions taken.

Thus a well-constructed report will give, as a minimum[10,4]:

−  a (brief) description of economic and market developments over the reporting period, to show management that the portfolio manager has been alert in his or her market analysis;

−  a description of the various positions taken during the month in response to market movements, to show management how the portfolio manager responded to his or her analysis of the market and what changes were made to the portfolio;

−  an analysis of the results of these actions and the returns made on the portfolio, to show how the profits earned relate to the positions taken.

Note that the purpose of the description of economic and market developments is to support the positions and returns analysis, not to provide an in-depth economic assessment.  Quite apart from the fact that the essence of the end-month note is timeliness, and overlong economic analysis will slow down its production, others in the central bank will be doing similar analysis anyway, and the reserves managers are unlikely to be the best placed to do this work.

A more detailed report might also include a forward-looking section to set out for management how the portfolio manager expects the market to develop, and how he or she is positioned to take advantage of this.  This may or may not include such details as scenario testing or stress testing, but should certainly include a list of all open positions (with, if possible, their current mark-to-market valuation).  A full position report might therefore read as follows:

A report such as this should be presented to management for each open position, with supporting text as required to explain the positions further or to explain how they fit into the portfolio manager’s overall analysis of the market and strategy for the portfolio.  Other elements of the report will depend more on individual circumstances and senior management preferences, but might include details of cash usage over the month, limit observances, risk positions, a breakdown of deals done with each counterparty, and so on.

As already stated in chapter 6, it is essential that hard figures such as positions, limit observances and returns are reported by the Middle Office.  This is to avoid any risk that portfolio managers could amend or hide uncomfortable news. Equally, judgmental elements of the report must come from the person responsible; ie  the portfolio manager.  This is to ensure that the reason behind every position or profit is given by the person taking it or making it.  The final report might therefore be structured as a body of text (analytic report, from the portfolio manager) with an

annex of figures (factual report, from the Middle Office accounting unit)[10,9].

  

External reporting. External reporting of a central bank’s reserves management operation serves two main purposes.  The first is Accountability:  the reserves are public assets and the central bank should account to the public for its management of them.  The exact method of making public the results of the reserves management operation (eg in the central bank’s annual report, or in a special report to parliament) and the degree of detail that is reported, is for each central bank to decide.  The most detailed reports will not only set out the size of the reserves but also explain the reserves management process, the benchmark used and perhaps even the returns due to active management against that benchmark.  Not all central banks will want or feel able to go into this much detail but as a minimum the central bank’s report should be sufficient to show that the reserves are all accounted for and are being managed according to established procedures[11].

The second purpose of external reporting is for Information, both to the IMF and others in the official sector and to other market participants.  The size of the reserves can show how much intervention a country has been doing, and can also provide reassurance to creditors on the creditworthiness of the country.  More detailed figures (as set out by the IMF’s current Data Dissemination Standards) provide valuable data for those in the official sector whose remit includes the maintenance of international financial stability.  Particularly for this purpose, it is important that the data is kept up to date:  confidence in a country’s external position can quickly be lost if regular reserves data releases start to be delayed or withheld from publication.

 

 

 

 

 

 

 

 

 

 

3. Management of China's foreign exchange reserves

3.1. China's foreign exchange reserves and SAFE

Chinese FX reserves have increased rapidly over the past, and such trend is likely to continue in the foreseeable future. As shown in Chart 1, in 1993 the size of Chinese FX reserves was only about USD 20.1bn, but this figure increased gradually to USD 165.6 bn in 2000, and this momentum has been accelerating since then. As a result, Chinese FX reserves amounted to USD 2.4tr. as of December 2009, accounting for 31.9% of global FX reserves in 2009 in comparison to 5.3% in 1995. Meanwhile, the Chinese FX reserves were approximately 48.9% of the country's GDP in 2009,[12, 3]

An increase in Chinese FX reserves has arisen from sizable surplus in both current and capital accounts, while in many other emerging economies, the increase mainly reflects surplus in the current account. Surplus in the current account in China is not surprising given that the country has been continuously running trade surpluses since the 1990s, and the level of the surplus got larger in 20065 and onwards – partly thanks to joining the WTO in 2001. It is reported that as of 2009 China exceeded Germany, becoming the world’s largest exporter. FDI inflow has also been far more than capital outflow in China, not only due to the strict capital control imposed by the Chinese government. Furthermore, since 2005 when the Chinese government announced to implement the new, more flexible exchange rate policy, it effectively induced increased amount of foreign capital inflow to China, and a large proportion of it was reported to be speculative international capital. How to manage the rapidly growing FX reserves has become a big challenge and somehow a burden for the Chinese authorities. Further, we consider how Chinese FX reserves are currently managed and the possible solutions available and conduct a detailed overview of the State Administration of Foreign Exchange (SAFE), i.e. the Chinese governmental agency in charge of administering China’s FX reserves.

SAFE is a subsidiary department under China’s central bank, namely People’s Bank of China (PoBC). However given the unique nature of SAFE, this institution is treated as equivalent to a vice-ministry agency. Therefore the head of SAFE has always been a deputy governor of the PoBC too. The legal foundation of SAFE relies on the Law on People’s Bank of China 1995, in which it is specified that the PoBC owns, administers and manages the country’s foreign reserves. In this context, PoBC delegates the tasks of administration and management to SAFE.

According to the Overview of Management of China’s Foreign Reserves (SAFE 2009), SAFE sticks to three principles concerning portfolio management, i.e. safety, liquidity and value appreciation. It is particularly stressed that maintaining the safety of China’s foreign reserves is the utmost task of SAFE[12,7].

Organisational structure. SAFE is headquartered in Beijing, and the head office in Beijing currently consists of eight departments.In addition, SAFE has 34 local bureaus and 2 foreign exchange administration centres across the country. Staff in local bureaus are mainly responsible for dealing with local needs on foreign trade and currency related issues, while management of reserves is centralised in Beijing. Meanwhile, SAFE has four offices abroad, i.e. in Hong Kong, Singapore, London and New York, where the offices in London and New York have their trading desks.

Investment strategy. Investment strategy of SAFE has been traditionally conservative, which is not surprising given that as a reserves manager, the main concerns are safety and liquidity. However, in the light of rapid accumulation of FX reserves in China, particularly over the past 10 years, SAFE has been becoming gradually more active. There are two main reasons for it[14,26].

1) Conservative management of the reserves creates huge opportunity cost6, if they continueto be invested in low-risk-low-return government bonds.

2) SAFE has been facing strong competition from its domestic rival, i.e. CIC over the past 2 years. It imposes significant pressures on the management of SAFE to increase return.

So far, SAFE mainly relies on its in-house team to manage and invest the reserves, i.e. staff from the Department of Reserves Management as noted above. However, since the late 1990s, SAFE has started to outsource some of its assets to external professional institutions, but its amount was reported to be very small. In addition, SAFE’s overseas offices have been able to conduct foreign trading. However, it is understood that staff in these offices mainly replicate portfolios of the head office in Beijing while benchmarked to pre-determined, strategic asset allocation. The exception might be the HK office, which has become more aggressive in recent years, i.e. investing in high risk and high return assets.

Asset allocation. China is one of the very few major countries which does not disclose portfolio composition data of its FX reserves. Different sources, however, have estimated such information. The market consensus is that by type of currency 60-70% of the reserves are invested in US dollars, 30-20% in Euro, and 10% in British pounds, Japanese yen and others (Zhang and He, 2008).

SAFE investment in the U.S. market. Table 1 gives the latest data for the top 3 countries in terms of holding value of U.S. securities.

It shows that as of June 2008 China as a whole held an amount equivalent to USD 1.2 tr. Of U.S. securities, accounting for 11.7% of the total U.S. securities held by foreigners, while Japan was still the largest holder by value .

Among the five broad groups of asset category, China has favoured long term debt, particularly long term treasury bonds and long term government agency bonds; the latter two almost accounted for 90% of the total investment, i.e. 43.3% and 43.7%, respectively as shown in Table 2. The heavy investment in US government agency bonds (e.g. mortgaged bonds) might explain why the Chinese government was so worried about the way in which the U.S. government bailed out Fannie Mae and Freddie Mac in 2009. In comparison, the largest holder (Japan) had relatively the same amount of U.S. Treasury bonds, but with much lower investment in the U.S. government agency bonds. Meanwhile Japan invested more than China in higher risky assets, i.e. equity and long term corporate debts. China is currently the largest holder of the U.S. treasury securities. As of November 2009

China held an amount equivalent to USD 789.6 bn of the U.S. treasury securities, in comparison to USD 757.3 bn for Japan [12, 16].

Recent trend of SAFE investment. For Chinese (FX) investment in the U.S. market, the pattern has undergone gradual changes over the past decade. As shown in Chart 2, in 2000 77.2% of assets were invested in U.S. treasury bonds, but in 2008 the proportion was reduced to 43.3%, almost a half reduction over the 8 years period. In the same period, investment in the U.S. government agency debt, however, increased significantly. It may reflect the changing behaviour of the SAFE, i.e. keep investing in safer assets but aiming to earn a higher return than that of the U.S. treasury bonds.

Although as of 2008 only 8.3% of the portfolio was invested in the U.S. equity market, much lower than what was invested in the U.S. treasury and agency bonds, the magnitude of the increase over the past 8 years was significant (i.e. almost an eight times increase). This observation again is consistent with the overall changing investment strategies of the Chinese government (i.e. SAFE).

SAFE has been recently receiving increasing attention both within China and abroad. Various factors have contributed to this.

In China • Rising assets and associated problems

AUM (asset under management) of SAFE stood at up to USD 2.4 tr. by 2009, and the rapid growth is expected to continue in the foreseeable future. In this context, the monetary authorities in China have been facing several serious problems, particularly with reference to monetary policy.

Firstly, PoBC has had to increase monetary supply in response to rapid accumulation of FX reserves. Chart 3 below shows that over the period between 1999 and 2009, Chinese monetary authorities witnessed rapidly increased credit creation due to accumulation of FX reserves. This in effect subjects monetary policy to foreign exchange policy in China, thereby creating inflexibility or difficulty in achieving intended monetary policy by PoBC[12, 17].