From the very inception in September 2008, when the crisis was just two weeks old (read link), when Sri Lanka had lost a little over 150 million US dollar LBO in particular had documented and predicted the crisis, step by step.
Fuss-budget had previously explained how a pegged central bank causes a currency crisis (read Thrift Column – Currency Crisis).
But as the latest currency crisis shows, it may be a good idea to go a little deeper into the problem, especially because it will help the public understand the severe pressures the economy will be put under as we try to recover from the crisis and collect reserves.
To refresh the minds of the readers let's just go back to exchange regime basics.
The most basic exchange rate regime is dollarization, where a country adopts the currency of a low inflation foreign central bank.
In the same family, is a currency board, where the monetary authority exchanges local currency for foreign currency inflows and outflows at a fixed rate. This is called a 'hard peg' and is maintained by daily interventions in the foreign exchange market which are not sterilized by off-setting interventions in money markets.
The domestic monetary base (reserve money) is 100 percent (or more) backed by foreign reserves. Therefore the monetary base is said to be created out of net foreign assets (NFA).
A currency board cannot manipulate interest rates in the system (because it cannot print money, cause high inflation and cause economic collapses, unless the anchor central bank to which the currency is pegged (like the Fed in the case of the dollar) messes up.
Hong Kong and a number of financial centres in small islands have such systems. People in currency board countries now have high living standards because the main weapon the state has to prey on the people - inflation by money printing - is not available under such a system.
Sri Lanka had such a regime until 1950 when a money printing central bank was created.
A pure floating currency does not need any foreign assets, though in practice such central banks may have foreign assets, usually as a legacy of a previous regime.
In a free floating regime the entire money supply is created by domestic assets, mostly government debt. These assets, (including the worthless junk that the Fed is now buying up), are called net domestic assets (NDA).
A floating central bank will not intervene in the foreign exchange market. Interventions are rare, and in all successful floating central banks, they are now only symbolic.
By the operation of a policy rate and open market operations, any foreign exchange intervention is automatically 'sterilized' and completely nullified in the money markets.
The sterilization happens this way. The central bank first sells foreign currency to a domestic player in exchange for its own previously issued currency. This contracts reserve money and can push up rates.
But through open market operations, enough money is automatically injected into money markets to maintain the interest rate via T-bill purchases, thereby 'sterilizing' the foreign exchange intervention (and also nullifying any appreciation of the currency).
An appreciation of the currency can only be brought about by raising policy rates, which will drain liquidity from the money markets.
Within the spectrum of fixed and floating, sits the third – and deadly - exchange rate regime. It is called a 'managed float' or a 'soft-peg'. Sri Lanka and any number of frequent customers of the International Monetary Fund have such systems.
Following the East Asian crisis and the Russian ruble hyperinflation many countries shifted to either floating rates (not intervening against currency movements) or towards currency board type arrangements.
In managed float countries, also called a dirty float, local money is created by a mixture of NFA (forex purchases) and NDA (T-bill purchases).
A pegged central bank imagines that while having a fixed exchange rate (and thereby supplying the entire money needs out of the balance of payments) it could also buy NDA and create some more money. But any NDA acquisitions will lead to higher inflation than the pegged currency, and currency crises may also occur.
A managed float country will try to target the exchange rate and interest rates (or domestic reserve money) at the same time. This is an impossible balancing act.
Inflow and outflow sterilization
When there is a capital inflow a soft-pegged central bank will intervene in the forex market and buy dollars to prevent the exchange rate from appreciating. This will increase NFA and reserve money by releasing rupees.
But because the central bank believes inflows will create inflation (in countries with foreign exchange controls this can happen) it sells treasury bills or its own securities and 'sterilizes' the money.
Such an action is a contractionary sterilization and we will call it inflow-sterilization. Foreign reserves will rise and even domestic interest rates can rise in such a situation.
In a currency board, the exchange rate will remain fixed in the face of an inflow, but rates will fall and ward off new inflows, while existing money may also leave the country to a higher yielding country because there are no exchange controls in such a regime.
But in a managed float, inflow-sterilization would push up rates and will encourage even more speculative inflows. This is what happened in Sri Lanka after Treasury bill markets were opened to foreigners.
When such capital leaves the opposite happens. To maintain the currency peg, the central bank will sell dollars. This should shrink the monetary base and raise interest rates. To prevent a contraction of the money base the central bill buy Treasury bills and print new money.
We will call this an expansionary sterilization or an outflow-sterilization.
Savvy foreign investors know that this is unsustainable and will quickly leave while the peg is maintained and a de facto forward exchange hedge is provided by the central bank.
If rates are cut while this happens, any remaining foreign money will also quickly leave.
To maintain the peg, the central bank will end up providing dollars for all transactions, including imports. In Sri Lanka a petroleum credit from Iran - which was even bigger than the total hot money volumes - also compounded the problem.
Outflow sterilization will swiftly cause a balance of payments crisis. Then the central bank will run out of reserves, the soft-peg will break and the country will go to the International Monetary Fund.
To correct the problem the exchange rate has to be floated and a process of outflow sterilization has to be reversed. Reserves are re-built through forcible inflow sterilization. This contractionary process can keep interest rates high and cripple an economy.
Reserve Pass through
To describe the extent of sterilization, monetary economists have come up with a measure called a sterilization coefficient.
Kurt Schuler, a US economist who authored the document 'Why currency crises happen' and is an authority on exchange regimes, has developed a concept called 'reserve pass through' which can easily identify currency regimes, based on the extent of sterilization.
Many central banks, like Sri Lanka's, claim to run 'floating' exchange rates but are in fact running different types of regimes at different periods with varying degrees of sterilization.
Under a currency board, reserve pass through will be 100 percent (sterilization coefficient of 1). All reserves collected will be passed on to reserve money and domestic players.
Under a pure floating exchange rate, reserve pass through will be zero. No reserves will be collected and none passed through. The local monetary base will be made out of domestic assets. There will also be no sterilization of inflows – coefficient zero.
The reserve pass through is measured by dividing the change in the monetary base by the change in net foreign assets, adjusted by the change in the exchange rate.
The reserve pass through curve here is developed by using the year-over-year changes in the monetary base and net foreign assets. It can also be developed using monthly changes but the curve is squiggly and harder to interpret for most people.
This curve was developed using published data of the central bank and using only the US dollar exchange rate for adjustments.
To get perfect results foreign assets must be adjusted by the exchange rate of each asset in the currency it is denominated if different types of assets are held. The exact institutional details relating to how the central bank interacts with the Treasury is also not available to this columnist.
As anyone can see, Sri Lanka does not have a floating exchange rate. The central bank is sterilizing all over the place, sometimes veering towards a currency board (near 100 percent reserve pass-through) and at other times going in the opposite direction.
Since September all the money has been created out of domestic assets (printed money) because the central bank has been a net seller in the forex market, and the money supply is behaving like that of a floating regime of zero reserve passthrough.
What does this mean?
People in a currency board country will get to use all the foreign inflows through the money supply.
People in a country with a floating exchange rate will get to use all the inflows through direct transactions in the forex market. In addition its money supply will be created with printed money (seigniorage). They will therefore have higher living standards (or be able to afford a bigger government).
A currency board country will invariably have a smaller government (without government-debt-backed money) or a more effective one like that of Singapore.
In a managed float country, whenever foreign reserves increase steeply, and the domestic money supply is controlled, capital will flow into US securities markets through inflow sterilization. US will benefit and the managed float country will lose out with higher interest rates and rationing of capital.
A managed float central bank will therefore impose severe costs and economic hardship in a period of inflow sterilization.
In a currency board country, any increase in reserves will also be accompanied by an increase in money supply and the local economy will not be hurt.
When outflow sterilization is done, by running down reserves, a temporary 'stimulus' can be given to the economy, but at the cost of inflation or at the cost of reserve collection in the future. This is why after the IMF comes, a severe contraction and economic hardship is seen.
A currency board country will have monetary related problems when the anchor currency country has a problem like the US is having now. A managed float country will have problems when the anchor currency country has trouble as well as when its own central bank messes up.
But an effective floating rate central bank can 'de-couple' itself from the rest of the world and ride out of external problems to a greater or lesser degree..
During a currency crisis, steep rise in domestic assets (T-bill stock) in a situation of severe liquidity shortages may not create the same inflation as in a domestic asset increase where there is excess liquidity.
If interest rates rise in a peg defence situation as it did in Sri Lanka, inflation will not be bad. This is somewhat similar to the phenomenon of capital inflows coming to finance traded goods as opposed to inflows into securities markets that can lower interest rates.
That is why the rupee proceeds of a 500 million dollar bond increased inflation to very high levels, but a run down of two billion dollars of reserves seems not to do to the same degree.
During the 2000 balance of payments crisis, inflation was not that high as in say 2006 or 2007. In a situation of capital flight and with even the monetary base contracting a little, some brakes are being applied.
That does not mean everything is ok. The currency crisis is still going strong. The problem of avoiding sovereign default (in a country where reserves have been traditionally used to pay off state loans) has been largely solved with the entry of the International Monetary Fund to the picture and a deal expected to be finalized soon.
Though gross reserves are 1.4 billion dollars by end-January, which includes fiscal reserves, a quick look at the central bank balance sheet shows a different picture.
Central bank's net foreign assets were already down to 1,482 million dollars by December when converted at the end-month exchange rate. By end January net foreign assets were only 1,142 million US dollars. An analysis of the monetary base and treasury bills holdings shows an even bleaker picture.
From January 30 to March 19, the monetary base fell from 263.8 billion rupees to 259.7 billion rupees. Treasury bill holdings increased to 193.1 billion rupees from 159.1 billion, which is a difference of about 34 billion rupees or 332 million US dollars.
Though the central bank said 200 million US dollars had been borrowed from Malaysia in February, we seem to have lost more than that already. Such borrowings will also create a short term liability for the central bank and put it in danger of being insolvent.
What the Iranian petroleum credit did to the fiscal side, the so-called 'swap' can do the monetary side.
The central bank's decision to limit repo deposits to 100 million rupees per bank will also not help the exchange rate. An LBO report on March 05 said why and said the US dollar is also likely to weaken against other currencies for the same reason. And it has started to happen.
One of the reasons the dollar strengthened last year was the excess reserves parked in the Federal Reserve which is similar to the repo window. Just like the dollar notes buried in coffee tins in the backyard kept the US currency strong during the Great Depression so did the dollars parked in the Fed.
In July 2008 excess reserves were under 2 billion dollars. By January excess reserves had climbed to 843 billion dollars.
By the first week of March they had fallen to 621 billion indicating that money was again being used by banks and the dollar is likely to fall, as intended by the Federal Reserve.
As a pegged country Sri Lanka will also get the 'benefit' of any dollar weakness.
The fall in commodity prices will be slowed or reversed and the rupee will weaken against other floating exchange rates as the Fed works overtime to print dollars, bring inflation back into the system to halt a slide in asset prices.
A US dollar currency board country, would just stay put and wait for the Fed to take action like Hong Kong does (subject to a drawn down of fiscal reserves). If necessary they can also borrow abroad.
As the crisis took hold, Singapore, which has a floating rate on paper, but targets the exchange rate on currency board principles, has let its currency weaken (from about 1.35 to 1.50) as the US dollar strengthened.
This reversed an earlier policy of gently nudging the Singapore dollar up to counter inflation flowing from the US dollar. (For more on Singapore's monetary policy read Thrift Column - Yawning Gap)
If Sri Lanka simply hard-pegged to the Singapore dollar we would have benefited both ways.
Sri Lanka on the other hand tried to fix hold the exchange rate with outflow sterilization and got into a mess. If the advantage of having a central bank is to have independent monetary policy, why did we try to maintain a fixed rate?
The US got out of the Great Depression by devaluing the dollar against gold. Britain did the same thing by lifting post WWI convertibility. The core causes of the Great Depression was the post war overvaluation of the Sterling (Read The Thrift Column - Deflationary Collapse)
There is a pithy Sinhala saying that describes the current situation succinctly. Yuddeta nethi kaduwa cos kotannader?, meaning if a sword is not available for war is it being kept to cut up jack fruit?
Sri Lanka has paid dearly for a having a 'managed float' soft-pegged central bank.
In the latest case we have had higher inflation than the US and Sri Lanka's exporters have been hit harder than our competitors, due to a soft-peg. By December 2008 the rupee was overvalued by 25 percent according to Central Bank's own real effective exchange rate index.
Sri Lanka tried to hold the exchange rate without an institutional framework to do so and lost most of our reserves. And we did it for puerile reasons like 'reducing the foreign debt burden'. That is so infantile we need hardly waste time talking about it.
While depreciation will help somewhat, reserve collection to rebuild and also pay back the IMF will impose severe contractionary pressure on the economy in the coming months. We will not only lose capital to USA on an absolute basis, but we will also be landed with higher than necessary interest rates in the recovery period.
This is what we have got for abandoning a currency board.
Higher interest rates than the rest of the world. Higher inflation than the rest of the world. More crises than the rest of the world.
The need for an IMF bailout shows that we among a bunch of countries to get into bigger messes than the rest of the world, in a time of global turmoil.