Asian Financial Crisis: The Battle to Defend Hong Kong’s Financial Stability

inSight

11 Sep 2019

Asian Financial Crisis: The Battle to Defend Hong Kong’s Financial Stability

During my nearly three decade career with the HKMA, I have had the unenviable experience of very close encounters with two major financial crises, namely the Asian Financial Crisis (AFC) which started in 1997 and the Global Financial Crisis of 2008.  In particular, during the AFC that rampaged across the region, there were attacks on Hong Kong, with speculators using a “double play” strategy to manipulate both the currency and stock markets, with the aim of breaking Hong Kong’s Linked Exchange Rate System (LERS) and profiteering from the stock market.  I was tasked to take command of a stock market operation.  While we were “fighting” the battle on the frontline, many critics at that time did not understand the reasons for our operation, which they believed was a betrayal of the fundamental principle of the free market.  So I would like to share with you my personal experience of this special episode in Hong Kong’s history.

A Perfect Storm was Brewing

Financial crises do not occur out of thin air.  Prior to the outbreak of the AFC, many economies in the region were already showing signs of vulnerabilities, including currency over-valuation, overheating of the economy, and excessive borrowing.  In particular, there were serious imbalances in the external sector, such as current account deficits, and high levels of external debt, as well as maturity and currency mismatch within the banking system.  Many international banks, especially those from Japan, were engaged in aggressive lending in the 1990s, leading to easy credit to the emerging economies in Asia.  All of this did not go unnoticed by speculators, including hedge funds, lurking in the wings.  They just needed the right moment to start preying on the yen and other Asian currencies.  The conditions were ripe for a perfect storm.

The Hedge Funds’ Menu

The shorting of the Japanese yen was the “main course” on the “menu” for the hedge funds.  They first borrowed yen from the Tokyo market at an interest rate of 3% or even lower for short-selling, and then bought 10-year US Treasuries at a 6% yield and leveraged up, or bought Russian US dollar (USD) bonds at a 10% yield and leveraged up.  These were regarded as safe and profitable trades.  The market generally believed that Europe, especially Germany, would never allow Russia to go into default because of concerns over spillovers into Europe.  Facing short-selling pressures, the yen steadily weakened from a high of around 80 per USD in April 1995 to 130 by the end of 1997 and further to 147 in August 1998.  Other Asian currencies, such as the Thai baht, Malaysian ringgit, Philippine peso, Indonesian rupiah and Korean won, were the “side dishes” for selection by the speculators for further short-selling.  

On 2 July 1997, the first business day for the Hong Kong market following the establishment of the Hong Kong Special Administrative Region on the previous day, I went to my office at 3 Garden Road as usual in the morning.  A senior official who was in charge of international affairs at the Bank of Thailand rang and told me that Thailand had allowed the baht to become free floating.  The baht then devalued sharply by more than 50% over the next few months.  This sparked off the devastating AFC, with the Indonesian rupiah, Malaysian ringgit, Philippine peso and Korean won falling one after the other like dominoes.  It was initially coined the Asian Currency Crisis but, as events unfolded, this proved a not very accurate label.  Whilst Asian currencies certainly were attacked and devalued sharply, the root causes and impact were much wider and deeper affecting, as they did, the entire financial system.

The Currency Speculators’ Playbook

Speculators would first look for some fundamental fault lines or vulnerabilities in a target currency and build up short positions during “quieter” or “calmer” periods in order not to raise any alarm, which could increase the cost of shorting. The reason was that a rise in the interest rates of the target currency would raise the cost of borrowing for the purpose of shorting and prematurely weaken the currency, thereby reducing the potential profits from the speculative shorting. 

Once the short positions had been more or less built, a campaign would be launched to generate pessimism or, better still, panic in the markets.  The intention was to provoke herding behaviour by international and domestic investors to dump the local currency for USD.  The speculators could adopt different means and channels, including the media, to spread fear in order to amplify the shocks and pressure on the target currency, thereby causing a sharp deterioration in sentiment as regards an already vulnerable economy.  Once the target currency collapsed, the speculators would take profit, squaring their short positions by buying back the target currency, at a drastically lower cost, to pay back the loans they owed.

All these speculative activities were highly leveraged as the shorting of the target currency could be achieved by selling the currency forwards, which has the same downward pressure on the exchange rate as selling the currency in the spot market.  The beauty of this shorting through forwards is that it would save the speculators the trouble of borrowing the target currency in the local money market and yet achieve the same degree of leverage.  Ultimately currency speculators did not have local currency and had to borrow from those who had.  Basically it was the local banking system or money market that was the source of liquidity or leverage for the currency speculators.  As hedge funds did not have big enough credit lines with the large local banks, they normally operated through, and hid behind, some investment banks or foreign banks, which were very eager, at least in the 1990s, to do business with the hedge funds.  So by and large, the currency speculators preferred, and were able, to conduct their highly leveraged play in relative obscurity (unless of course their hubris got the better of them and they felt the need to boast about it publicly).

Attacks on Hong Kong

While Hong Kong was in a relatively stronger position than its Asian peers prior to the AFC, there were obvious fault lines: (i) there was a huge property bubble; (ii) households were heavily indebted mainly due to mortgage borrowings; (iii) corporates, especially property developers, were over-geared; and (iv) prevailing trade deficits were running at around 3% of GDP, meaning that Hong Kong imported more than it could export and spent more than it could earn.  This indicated a clear overheating of the economy and a loss of competitiveness of the HKD versus the USD.  So Hong Kong’s LERS became an obvious target.  The not-too-big and not-too-small size and high liquidity of the HKD markets, coupled with total freedom of movement of funds in and out, made Hong Kong a very attractive target.

The first wave of attacks occurred in August 1997.  Then, during the week of 20 October, short-selling in the HKD intensified.  The defence, under the design of the Currency Board System, would be that the shorting of HKD would bring the Aggregate Balance (which was at normal times HK$2 – 3 billion) to a lower level or even negative territory, causing HKD interest rates to rise sharply, thereby raising the cost of shorting HKD through borrowing or forwards.  And it did.  Shortly after market opening on 23 October – “Black Thursday” – overnight HIBOR once shot up to a shocking high of almost 300%.  With the cost of shorting becoming untenable, shorting activities subsided with the outflow pressure easing somewhat.  Overnight HIBOR then fell back to 5-6% a few days later, but one-month HIBOR still stayed above 10%, which was significantly higher than the level before the attack.  It was clear that an ultra high HKD interest rate would significantly alter the economics of shorting, but it was also obvious that such high funding costs would do a lot of harm to the real economy and the financial system if they should last for a prolonged period of time. 

A Short Respite

While Hong Kong seemed to have fended off the speculative attacks in October 1997, the other Asian economies fared very badly in the meantime.  During the crisis period, the Thai baht devalued by 56% and the Indonesian rupiah by 85%, and both countries went to the International Monetary Fund (IMF) for help.  South Korea also went for an IMF Programme after the Korean won devalued beyond 1,000 per USD and it had almost depleted all of its foreign currency reserves while Malaysia imposed exchange control.  It is worth noting that the Japanese yen, which was the “main course” on the menu of the currency speculators, also experienced strong devaluation pressure.  The market was very pessimistic about Japan, referring to the “implosion” of the Japanese economy.  The yen dropped from around 110 per USD in mid-1997 to around 147 within just 12 months, and anecdotally foreign exchange dealers were, in mid-1998, predicting 170 or even 200 by the end of that year.

In Hong Kong, there was a period of relative calm in the HKD exchange rate until August 1998.  However, there were some trends that were discomforting.  First, the one month HIBOR, which was the key benchmark for the funding costs of banks for their mortgages and other loans, remained at an elevated level of over 10% in the few months after October 1997.  This put enormous pressure on banks to raise their prime rates by several percentage points, which inevitably would deal a further blow to the already collapsing property market (which dropped by almost 50% in 12 months from its peak in 1997).  While the banks held back the increase in prime rates, they could not hold for long if HIBORs did not ease back.  Secondly, the stock market was steadily dropping accompanied by a significant fall in the trading volume, with daily turnover shrinking from an average of HK$15 billion in 1997 to just HK$4 billion in July 1998.  At the same time, the Hang Seng Index (HSI) futures market saw its total open positions gradually rising from 59,000 contracts at the end of 1997 to 98,000 contracts at the end of July 1998.

It was not entirely clear to us at that time what was going on, but there was a clear sense of unease with a hunch that a new wave of speculative attacks, likely to be different and on a larger scale than the October 1997 episode, was about to hit Hong Kong.

The “Double Play” and Our Fight Back

By around August 1998, with the pressure on the HKD steadily increasing, we realised that the attacks had started.  Short-selling had intensified with several investment banks taking the lead.  At the same time, there were many media stories that the Hong Kong stock market and the LERS were bound to collapse.  The sentiment in the financial markets and amongst the general public was so bad that it was clear the optimal time for the launch of a full scale attack had come.  It was also becoming obvious that the speculators, taking a lesson from the October 1997 episode, had changed their strategy in two major ways: (a) they had built up major short positions in the stock market as well as the HSI futures market, aiming to profit from a sharp fall in both markets when the currency attack began, with the resultant sharp rise in interest rates; and (b) learning from the inhibitive high cost of shorting back then, they had “prefunded” themselves with HKD (believed to be to the tune of around HK$30 billion) by borrowing in the money markets during the “quiet” months before their attack.  So a rise in HIBOR during the attack should do them little harm.  This was a clever strategy, which we dubbed the “double play”, as it circumvented the interest rate defence mechanism of the Currency Board System and created synergy with a mutually reinforcing pressure on both the currency and stock markets.

We considered that the stability of Hong Kong’s monetary and financial system would be under very serious threat should the speculative attack be allowed to continue.  But how could we counter the double play?  We decided that we should use market means.  Under the Currency Board System, the HKMA was already intervening in the foreign exchange market to maintain the HKD exchange rate at 7.8 to one USD.  As for the stock and HSI futures markets, we would deploy the Exchange Fund to counter the manipulation of the speculators.  It was a painful but necessary decision.  To do nothing was not an option as there would have been dire consequences, as in the case of our neighbours, if Hong Kong’s financial system was to collapse under the speculative pressure.  Ours was a highly unorthodox move that was not, we believed, within the contemplation of the speculators.

D-Day

On Friday, 14 August 1998, the CEOs of the three largest stock brokers in Hong Kong were invited to the China Club in Central to attend a breakfast meeting called, at very short notice, by the Financial Services Bureau of the Hong Kong SAR Government.  When they arrived, they were surprised to see me, and me alone.  Hitherto the HKMA had had no dealings with the stock brokers in Hong Kong, as the Exchange Fund did not make any investments in equities at all.  I asked them to finish their coffee and switch off their mobile phones, and then took them to the HKMA office.  They were told, in strict confidence, that the Government had decided to intervene in the stock and futures markets to counter the double play.  They would need to go back to their offices and open stock and futures trading accounts for the HKMA immediately, as we would soon be starting the operation on the very same day.  That was the beginning of the stock market operation.  The HSI reversed its downward trend and rebounded by 564 points, or 8.5%, that day.  To check on the confidentiality of the operation, I asked my dealing room staff to enquire with market players during the day as to the reason for this rebound.  None came back with any hints of the Government intervention.  Of course, the news broke when Mr Donald Tsang, then Financial Secretary, Mr Rafael Hui, then Secretary for Financial Services, and Mr Joseph Yam, then Chief Executive of the HKMA, gave a press conference after market close that day.

The Stock Market Operation

The operation lasted for ten trading days, ending on 28 August 1998.  It was an unprecedented and hazardous endeavour from day one.  Prior to that, the Exchange Fund had not directly invested in or held any equities.  So we did not even have any stock trading account with anybody anywhere.  Besides, the HKMA dealing room was designed to trade only foreign exchange, forwards, interest rates and bonds.  So we did not have the trading facilities for stocks and HSI futures.  Obviously, the stock market operation was hugely market sensitive and we needed to maintain absolute confidentiality until the launch day.  We could only involve very few people within the HKMA, who were sworn to secrecy in the preparation for the operation.  I convened a small team of “operatives” and commandeered Ms Amy Yip, then Executive Director (Reserves Management), to man the “trading” cum “war” room for the operation.  We set up half a dozen telephone lines with make-shift recording facilities for orders to be made to our stock brokers.  Our mission was clear: to stand in the market and buy and thereby prevent the speculators’ manipulative trading strategies from causing excessive falls in the stock and HSI futures markets, which would destabilise our financial system.

We were only interested in, and only dealt in, the HSI constituent stocks and the HSI futures.  Selling pressure in the cash and futures markets was very heavy but the shorting activities varied from stock to stock and from day to day.  Our aim was to counter the fall in the HSI with the minimum amount of “ammunition”.  So our buying strategy had to be tailored and adapted to the prevailing conditions in the market.  As our market operation intensified, we used several more brokerage houses to handle our orders.  However, in all cases, we would not discuss the trading strategies with the brokers in advance: they only executed orders as and when received from us.  This approach worked well as it helped reduce any potential risk of front running, even though the market knew full well the HKMA was in the market all the time. 

On 28 August, the last day of the operation, selling pressure from the speculators, who had now become wary of the prospects for the success of their strategy in Hong Kong, reached an unprecedented peak.  The trading volume was a historical high of HK$79 billion, with the HKMA almost being the only buyer in town.  As for the HSI futures, the total outstanding open positions were also a high of over 150,000 contracts.  The HSI finished the day at 7,830, a rise of 18% from the level when the operation began, instead of the 4,000 level which we believed the speculators had been aiming for.

The Short Squeeze that Did Not Materialise

Before the end of the operation, we contemplated whether the short-sellers could themselves be short-squeezed.  While it was hard to know, we reckoned that the bulk of the HK$79 billion worth of stocks sold to the HKMA on 28 August were sold short.  This logically meant that the short-sellers would need to borrow stocks to settle their trades.  The usual stock lenders in such circumstances were long-term investors who have stocks lying idle in their securities accounts and who have joined their custodians’ stock-lending programmes to obtain some (relatively small) fees on their large stock holdings to help defray custody and other costs.  In theory, if speculators made use of these programmes to launch a large-scale shorting of stocks with the aim of depressing stock prices, it would ultimately work significantly against the interests of the stock owners.  We made this point loud and clear, hoping to catch the attention not only of those owners of large amounts of HSI stocks that we could identify but also of the custodians.  We also sought agreement from the Clearing House for the Hong Kong Stock Exchange that it would strictly enforce the T+2 settlement rule in the Clearing House Rules.  This would mean that if a short-seller failed to deliver on T+2, the Clearing House would undertake a compulsory buy-in in the market at prevailing prices on behalf of the short seller who would then be liable for any losses.  If both stock-lending supply and timing were tight, the likely result would be more difficulty and potentially higher costs for short sellers to cover their shorts.  Ultimately, however, we were not able to secure strict application of the Clearing House Rules’ T+2 deadline for the trades transacted on 28 August and the short-sellers were allowed to settle on T+5.  

The Unwinding of the Speculative Shorts

Meanwhile, the Russian economy had been running further into trouble. On 17 August 1998, the Russian Government devalued the ruble, defaulted on domestic debt, and declared a moratorium on repayment of foreign debt.  Then on 2 September, the Central Bank of the Russian Federation decided to float the ruble freely, which quickly lost about 60% of its value in the ensuing week.  As a result, those holding significant long positions in Russian government debt faced severe losses and this was a contributory factor in the eventual collapse of the highly leveraged Long Term Capital Management.  The banks that provided the credit facilities for these highly leveraged plays cut or withdrew their lines en masse.  This led to a sudden and massive unwinding of the short positions held by the speculators in the Asian markets.  Notably in this unwinding rampage, the yen rose sharply by over 15% against the USD in just a few days in early October, indicating the magnitude of the short positions held by the currency speculators.  Similarly, the short positions held by them in Hong Kong and other Asian markets were also unwound.  This was reflected in the return to normality of HKD interest rates, rebound of the HSI, and reduction of open positions in HSI futures.  A battle with not a smoking gun in sight thus ended.  We then had to face the difficult task of rebuilding market confidence.

Addressing International Criticisms of the Market Operation

When the market operation was launched, there was widespread criticism of this unorthodox move.  The criticism centred around the thesis that “No Government has ever succeeded in jacking up the market”.  It was regarded as a betrayal of the free market principle for any government to intervene in a stock market to prevent it from falling.  But what the critics had missed was that Hong Kong was, and still is, a staunch supporter of the free market principle.  However, classic economics theory does not address the question of what one could or should do when a mid-size market is faced with a potential systemic collapse of its monetary and financial systems caused by the deliberate manipulative actions of a few very large players who were able to push the market well beyond what could be justified by fundamentals.  In the 12 months after the market operation, senior officials, including the Financial Secretary, Secretary for Financial Services, Chief Executive of the HKMA and myself, travelled around the world explaining why we had to do what we did without abandoning the free market principle that we had always cherished.  Many of these critics, in later years, particularly after the advent of the Global Financial Crisis in 2008, came to appreciate the point on the need for government intervention to maintain stability in the event of severe market failure for whatever reason.  Intervention from governments and the public sector during the Global Financial Crisis took a variety of forms, including bank bail-outs and asset/deposit guarantees, and reached unprecedented levels, as various jurisdictions did whatever they had to in order to preserve their financial systems and the functioning of their economies.

 

 

Norman Chan
Chief Executive
Hong Kong Monetary Authority

11 September 2019

Latest inSight
Last revision date : 11 September 2019