OPTIONS FOR DEVELOPING THE CHINESE FX MARKET

Why are foreign exchange markets important?

For such an important trading country as China - a country whose share of global trade flows is increasing rapidly and seems destined to continue to do so for many years - the exchange rate is an extremely important price. And the market for foreign exchange, therefore, is vitally important. This is not the same as saying the exchange rate should be allowed to fluctuate freely. That is a monetary policy issue. But whether the exchange rate is fixed or freely floating or somewhere in between, it is important to have a fair and reliable market.

That is, I hope, a non-controversial statement. However, the current debate - at least outside China - about the yuan trading regime is often confused, misguided and at times malicious. I often say that exchange rate policy is more about politics than economics, and it is clear from the way this particular aspect of Chinese economic policy is being discussed in the United States and some other countries that politics is indeed at the heart of it. But the Chinese officials are correct to emphasize that foreign exchange and capital account policies should be determined by local conditions; that the pace of liberalization should be dictated by the pace of reforms in other areas and by the capacity of financial institutions and other parties to cope with the changes. I would not agree that many foreign commentators seem to equate greater flexibility of the exchange rate with capital account liberalization, when the two are distinct issues. If there is one thing that has been learned from two decades of emerging market crises, it is that capital account liberalization is best done carefully and with due regard for financial system soundness.

Such pleas for caution all too often are an excuse for inaction, but nobody who has followed Chinese financial reforms can argue that that is the case here. Instead, there has been a careful and deliberate but nonetheless significant progression towards a more liberal financial system with greater play for market forces to allocate capital. And indeed, the pace of reform seems to be accelerating, not slowing, as the recent liberalization of the use of the yuan in Hong Kong illustrates.

This need for caution argues against the kind of radical changes in policy being pushed by some commentators abroad. Those arguments are, moreover, in our view deeply flawed. To argue from the perspective of one bilateral trade relationship that the overall policy with regards to the balance of payments and exchange rate determination is inappropriate is simply bad economics. China may reach a USD125bn trade surplus with the US in 2003, but this country runs a trade deficit with respect to most other countries. It is difficult to argue that the RMB is significantly undervalued when China is running a current account surplus of only about 1% of GDP. EøR

What is needed in this debate is some microeconomic analysis. The growing surplus with the US is largely a reflection of changing investment decisions by US firms and their suppliers. More than half of China's exports are produced by foreign-owned firms. Three of the eleven largest exporters in 2002 were American firms. As the producers of final goods, whether it be consumer goods or sophisticated capital equipment, move to China attracted by lower production costs and the lure of a large domestic market, the largest consumer of traded goods - the US - is naturally going to see its deficit with China grow. So too is the European Union's deficit with China.

But this trade deficit is partly offset by capital flows back to the US (and Europe). First, US EøR companies who invest in China repatriate their profits back to the US, so the trade deficit with China overstates the US' current account deficit with China. But except for the mercantilists in the US Congress most people view the availability of cheaper goods as a positive - it makes higher value added goods more affordable, helps to restrain US inflation and keeps US interest rates down. The latter effect is reinforced by what the Chinese authorities do with their surpluses - they reinvest them back in the US. China, along with the other central banks in Asia have, by virtue of their purchases of US assets as part of their foreign exchange reserves, financed about two-thirds of the US current account and fiscal deficits in recent months. Hence, the very policy that politicians in the US complain about is preventing a sharper selloff of the USD and helping to prevent US long-term interest rates from rising further than they already have.

China’s foreign exchange policy is vitally important to the global economy. And so decisions about how to develop the market have implications not only for the future for this economy, but for the broader world.

Considerations in market design

What is it that we want in designing the foreign exchange market? Naturally, we want accurate pricing, efficiency and security. That is, we want a fair price, at minimum transactions costs, and we want to be sure that when someone commits to delivering a certain amount of currency at a certain date at a certain price, that we will get exactly that.

 EøR

Once we enter into the realm of market microstructure or design, there are no hard and fast rules. But some stylized facts do seem to hold. For pricing to be accurate - to reflect the "market" - access to the trading mechanism needs to be as broad as possible. When many potential suppliers or demanders for a product are excluded from the market, or if information relevant to the market - especially prices - is not made freely available, it is difficult to be confident that the resulting price is "fair." EøR This is a vital feature in market design.

Price discovery

Opening up the market to as many potential traders as possible also helps to keep transactions costs low. Where bid/ask spreads are negotiable, they should be narrower when a bank has other traders or dealers who may willingly take over a position at the current market price. In the foreign exchange market, however, there is evidence that there can be "too much of a good thing."

Almost everywhere else in the world, the foreign exchange market is an inter-dealer market. A client phones his bank to place an order to buy or sell a certain amount of foreign currency at a certain date or under certain conditions. The bank may then itself buy or sell the currency from or to another bank. More often than not, both the client and the bank will phone a few different banks to get the best price. Importantly, foreign exchange is not usually traded on an exchange or through an automated dealing mechanism. The market has operated like this for decades. But that isn't necessarily the most efficient way to design the market.In an inter-dealer market, higher turnover can be associated with lower information content in market prices, higher bid/ask spreads and greater volatility. In the global foreign exchange market, as much as two-thirds of trades are between dealers. These reflect in many cases risk management imperatives - a bank that executes a large order for a client may have too large an exposure to retain on its books. It will then sell parts of that position to other banks. Some transactions with clients require a larger number of transactions by the dealer in order to lay off the risk, for example, from a forward sale of foreign currency. So risk management often means that one trade with a client gives rise to many more trades between dealers to spread the risk around.

But it appears that in dealer-dominated markets, a significant amount of turnover reflects trades undertaken simply for price discovery purposes. Dealers will offer to buy and sell foreign exchange simply to test the willingness of other banks to trade at current market prices. Since the intent is not to add to risk positions at the end of the day, these are reversed immediately, but the result is a dizzying number of transactions apparently with no underlying "real need" but whose effect is to ensure that the market price really is the market-clearing price.

But that need not necessarily be the case, because this price discovery mechanism can actually raise the cost of finding out what the correct market price is. Because of information asymmetries among dealers, it can also be the case that this pattern of rapid turnover can increase bid/ask spreads by reducing the information content of each trade. If it is really the client with a real need to buy or sell foreign currency who provides genuinely important information to the market, as their importance in the market falls due to the rise of unrelated inter-dealer trading, the signal the client's trade provides is diluted.

A trading platform in which price discovery is more transparent - an electronic trading system, for example, as you have here with CFETS - may not only be more efficient, it may also be the future for the rest of the industry. The Bank for International Settlement reports that up to 70% of spot foreign exchange trading in the major currencies may now be traded on electronic broking systems, up from only 10% in 1995. Such systems not only are more reliable - reducing the possibility of human error - they offer more transparent pricing and lower transactions costs. Our own experience mirrors this trend.

One need only consider the transformation of the futures trading industry in recent years - the demise of the trading pit - to see how the greater transparency and efficiency of electronic trading can erode even the most venerable trading institutions. Even the NYSE is threatened by electronic trading systems. It is not unreasonable to suppose that over time the inter-dealer foreign exchange market will be largely replaced by electronic brokers.

Settlement and credit risk

An inter-dealer market also has the disadvantage of embedding credit risk in almost all transactions, which therefore affects market prices. Any transaction for non-simultaneous delivery involves credit risk. Much time and effort has been invested to reduce settlement risk in the foreign exchange market, moving towards shorter settlement times and payment-versus-payment settlement systems.

But settlement risk is closely related to credit risk, which for emerging markets is probably an even bigger concern. Credit risk can be a significant barrier to entry, and certain creates an "un-level playing field" in which different participants may be charged different prices to reflect their relative credit risks. The greater the risk, the higher the transaction cost, and the lower the quality of the resulting price. This isn't unfair trading. In a market in which dealers have to take credit risk against their counterparties even for a brief period of time, they will want to be able to charge for that credit. In the extreme, higher-risk banks or firms may be excluded from the market altogether because other dealers will not extend credit lines to them.

It may be preferable therefore, to consider an alternative market architecture in which this counterparty risk is mitigated. Here again, CFETS has provided an important improvement over the traditional market architecture, by removing credit risk from the transaction. The risk doesn't disappear, of course, but it is shared and mitigated in a way that removes this barrier to entry. By taking credit risk premia out of observed market prices, CFETS improves transparency and access to the market.

So, with the benefit of many years' experience elsewhere to draw on, the Chinese authorities have designed a trading infrastructure that not only fits local conditions, but also improves on market design elsewhere, and probably fits with the direction in which the global foreign exchange market is moving. Having an electronic broker automatically in real time matching orders to buy and sell foreign exchange may well be a more efficient price discovery mechanism than an inter-bank or inter-dealer market. And having the exchange guarantee each leg of the trade not only reduces the risk to participants, it allows banks with weaker credit histories to participate in the market.

A market for what?

As we consider options for developing the foreign exchange market, it is important to step back and consider what it is that we are trying to create. There is in principle an extremely broad range of potential market participants, each with their own interests, but we can assume that these can be summarized as follows: there is interest in having a market in which to exchange RMB and foreign currencies today, for delivery as soon as possible; for delivery at some fixed date in the future; and for delivery according to some uncertain event or condition. In market terms, these are a demand for spot delivery, for a forward or futures market, and for currency options.

The first step is the spot foreign exchange market, not only because of its broader appeal, but because spot prices are needed in order to value other types of foreign exchange contracts. On CFETS, four bilateral exchange rate contracts are currently traded. But others are possible. Indeed, the central bank has authorized banks to deal in bilateral G7 currency pairs - e.g., USD/JPY - but for many banks, an inability to access credit from dealing banks has prevented them from being able to enter that market. Including these currency pairs on CFETS seems not only a natural extension of the current suite of products, but also allows the central bank's policy to be put into effect.

The next step is to introduce contracts for forward delivery. In an electronic broker market, these would actually be futures instead of forwards, which are bilateral, inter-bank contracts. A futures contract implies exactly the same economic exposure to foreign exchange rates but without the same credit risk - a contract is purchased with full payment up front or with a margin, but the amount of counterparty exposure is reduced.

Introducing forward or futures contracts is more urgent than it often appears. Any effort on the part of banks or corporations to manage their foreign exchange exposures requires an ability to trade today for future delivery. Such contracts, therefore, are integral to banks' and corporates' efforts to manage (i.e. reduce) their foreign exchange risk. I would put the introduction of currency futures contracts high on the list of near-term priorities for market development.

Finally, experience has shown that there is strong demand for optionality in foreign exchange markets - the ability to make transactions conditional on future events. The most common, of course, would be call or put options on foreign exchange, giving someone the right to buy or sell foreign currency if its price reaches a certain specified level. Giving banks and commercial clients the right to buy options - as opposed to the right to write options, which is a far riskier thing - would be a valuable addition to the market.

However, I would not place this as a priority for market development for two reasons. First, to have any confidence that options prices are fair one needs a long history with an efficient spot market in the underlying currency. Options trading, therefore, should lag the development of a liquid spot market. Moreover, when the exchange rate does not fluctuate, as is the case with the RMB, one is unlikely to develop a liquid options market that doesn't simply replicate the forward market in the economic risks that can be priced.

Developing exchange traded options is also less of a priority in my opinion because the optionality that is demanded by one client is unlikely to be the same as that demanded by another since the underlying risks are likely to be very different. The ability to enter into tailored options contracts that specifically target the underlying risks that a client wants to mitigate is a clear advantage of an inter-dealer market over an exchange-traded market. Note, for example, that there is very limited trading in exchange traded options even in the major currencies compared to the OTC options market.

Links between FX and other financial markets

Development of the foreign exchange market goes hand in hand with development of other financial markets, especially fixed income markets. Foreign exchange and money markets are integrated to a significant degree - almost perfect substitutes for each other in completely open capital account regimes - through the covered interest parity condition: interest differentials between countries are reflected in forward (or future) foreign exchange premia. Indeed, in many countries the foreign exchange market is actually the source of the most reliable short-term local currency interest rate.

This is a clear advantage of the market architecture here in China, in which CFETS is not only the trading platform for the foreign exchange market but also for the money markets. It also suggests that we should be thinking not only of how to further develop the foreign exchange market, but also how to further develop the money market using the same trading platform.

 EøR

Conclusion/summary

China is the world's fifth largest trading country, and its share of global trade is rising fast, for reasons that are far more complicated and interesting than simply the level of its exchange rate. But as foreign currency transactions become more important to the Chinese economy, it is vital that the market in which currencies are traded in China is as efficient as possible. Certainly, among the many preconditions to capital account liberalization or even a flexible exchange rate regime I would include having a well developed market in which parties can actually trade currencies. In my opinion, the Chinese authorities have already chosen a market infrastructure that has many advantages over foreign exchange markets elsewhere. By adopting an electronic broking system in which credit risk is mitigated by the central clearinghouse the authorities have adopted the very market infrastructure that the international markets are heading towards. While an interbank market may have advantages over an exchange in the options market, CFETS offers significant advantages over an inter-dealer market in spot and forward foreign exchange.

In my view, the next steps should be to broaden the range of currency pairs traded on CFETS, to include the major G3 currency pairs - USD/EUR, USD/JPY, EUR/JPY - and then others as they are permitted. By creating a transparent and level playing field for trading foreign currency CFETS offers the potential of a highly efficient market in which to determine this most important price - the price of the nation's currency.

The next step - and I would view this as a matter of some urgency - is to provide the key risk management instrument that is needed in a foreign exchange market, the ability to trade currencies for future delivery. In the mature markets, two-thirds of foreign exchange market turnover is in the forwards and swaps markets, which can be replicated here by futures and spot contracts. By giving trading companies the ability to lock in prices at which they transact at dates beyond the next couple of days, CFETS can make an important contribution to helping firms and banks to manage their foreign currency exposures as the authorities move towards capital account convertibility and greater currency flexibility.

Finally, I would note the synergies that exist between trading in foreign exchange and local money markets. I would expect that as the foreign exchange market is developed, this will facilitate the development of the local money market and interest rate liberalization.