Sri Lanka’s severe monetary instability, the obsession with imports, the trade and current account deficit, despite ending a 30-year war present a graphic example of how Keynesian ideology, which is not based on logic, reason, or economics, can devastate countries at peace.
Sri Lanka’s Latin America style central bank is borrowing dollars from Bangladesh which has a much better overnight rate framework – even better than the Reserve Bank of India – which has allowed the country to have over 10 years of monetary stability since the 2008/2009 Federal Reserve bubble collapse.
Friederich Hayek once said that Keynes appeared to understand very little of the history of monetary instability of his own country. Keynes was a “man with a great many ideas who knew very little economics,” Hayek said.
“He knew nothing but Marshallian economics, he was completely unaware of what was going on elsewhere. He even knew very little about the nineteenth century economic history. His interests were very largely guided by aesthetic appeal. And he hated the nineteenth century and therefore knew very little about it. He was an expert about the Elizabethan age.
“Even within the English tradition, he knew very little of the great monetary writers of the nineteenth century. He would know nothing about Henry Thornton (anti-bullionist), he knew a little about Ricardo of course – the famous things. But he could have found any number of antecedents of his inflationary ideas in the 1820’s and 1830’s. And when I told him about it, it was all new to him.
“He took it for granted that Marshall’s textbook contained everything one needs to know about the subject. There was a certain arrogance of Cambridge Economics about it. And they thought they were the centre of the world. And if you have learned Cambridge Economics there is nothing else worth learning.”
Money printing in the form of provisional advances, outright Treasury bill purchases to defend Treasury yields compounded by a surrender requirement, which is worse than a suspension of convertibility, have placed the country on a seemingly irrevocable path of monetary meltdown.
The Mercantilist obsession with the trade deficit (in Keynes’ time it was called the ‘commercial balance’ by classical economists who tried to reason with him) and neo-Mercantilist obsession with the current account deficit, are all signs of the problem of monetary policy neglect.
In 2020, authorities claimed that imports fell due to controls, this column among others said that it was due to contraction in credit and consumption. Further it was predicted that imports would recover once credit recovered and problems would become worse as the U.S. was printing money and would drive commodity prices higher and any subsidies re-financed with excess liquidity will only make the problem worse. In the absence of central bank credit (printed money), there cannot be a forex shortage at a peg of Rs200 to the U.S. dollar.
Mercantilism vs Reason
Mercantilist obsession with the trade deficit and neo-Mercantilist obsession with the current account deficit, are all signs of the problem of monetary policy neglect.
WHAT IS A PEGGED EXCHANGE RATE?
In a peg, the monetary authority will buy and sell foreign exchange in unlimited quantities at the pegged rate. A purchase of dollars by the central bank from an exporter will increase excess liquidity in the banking system by inserting new reserves to the bank and the dollar seller. The account of the customer who sold the dollars will be credited with the newly created money. This action will increase reserve money and overnight rates may fall.
After the central bank buys dollars to create liquidity the bank can not only loan the rupees it gets as loan repayments and deposits, but also the newly created excess liquidity from the central bank dollar purchase – or surrenders.
If the customer does not spend the money it will be loaned to another. The borrower may invest it in a hotel or house, thus, triggering imports. Even if the person bought a domestically produced product, in the next round the money will hit the forex market. Import substitution therefore will not stop the problem of forex shortages even if there were no imported inputs in it.
When the new money hits the forex market, the central bank will sell the dollars it bought in the first place and the reserve money will contract once more. The bank of the customer who bought the money will see a reduction in rupee reserves.
Following the dollar sale by the central bank, the volume of loans the importer’s bank (and therefore the banking system) will be able to grant will be less than the total of loan repayments and net deposits it received. Overnight rates will tend to increase.
This is why in countries where there is a wide policy corridor (stable peg) or there is no policy rate at all (currency board) the exchange rate will be stable or be permanently fixed. Sri Lanka’s rupee was fixed from 1885 to 1950 through a currency board. Since these central bank’s piggy back on the US rate, they will also not collapse when the Fed tightens policy as Sri Lanka does all the time.
RESERVE PASS THROUGH
It can also be seen that in a pegged exchange rate, all excess exports and remittances inflows exceeding daily import demand as well as days when inward remittance are overtaken by outward flows, one will pass ‘through’ the reserve money altering its size and the overnight rates.
Dollar inflows will make the reserve money bigger and new outflows make it smaller every day because the exchange rate is fixed, altering excess liquidity (the aggregate balance of the banking system).
The time the excess liquidity remains in the system and is turned into credit or deposit withdrawals, is the timing difference between the inflow of the dollars and the time it takes for someone to spend the money or borrow and invest. When a private borrower takes the money and builds a hotel or apartment, if the government borrowed the money via a Treasury bill and repaid the loan, the effect will be the same.
The exchange rate will be fixed, reserve money will narrow and interest rates will go up. In a consistent peg, the reserve pass through is 100 percent of dollar flows or 1 to 1. Therefore there are no forex shortages even if the government repays a loan since the sale of a Treasury bill to a bank or other customers will crowd out another borrowing.
The interest rate going up and down is a consequence of how the crowding out will happen, but there is no need for a depreciation of the currency, since reserve money change is the consequence of an inflow or outflow of dollars.
It can be seen that in a peg there is no possibility of dollar outflows exceeding inflows in the short term it takes for liquidity to turn into credit, which seems to be around 4 to 6 weeks in Sri Lanka, based on observations of this columnist.
If there is a persistent fall in credit, dollar outflows will fall and the central bank will be able to build up forex reserves and excess liquidity will build up. Though the daily changes in reserve money, dollar outflows exactly match outflows and the timing difference is reflected in the change in reserve money.
This is why in Vietnam and Bangladesh, as consumption fell foreign reserves went up. Overnight rates in Bangladesh fell 90 basis points and about 20 basis points in Vietnam. Such regimes have fairly credible pegs.
In a soft-peg, by selling down central bank held securities it is possible to build up forex reserves in excess of the reserve money. When a security is sold to a bank or a customer, an equal amount of rupees will not be available for consumption or investment.
Keynes taught at Cambridge University. Alvin Hansen was at Harvard. They had the tools and the power to mislead a generation of economists – the lost generation – for which countries like Sri Lanka are paying the price to this day
FLOATING RATE
In a floating rate, the reserve money is not created by the sale and purchase of dollars. In a floating rate, reserve money is created by open market operations through the purchase or sale of dollars to maintain a specific policy rate. Reserve money is not altered by dollar flows. Dollars flow in and out, outside of reserve money. Since there are no interventions in the forex market, and there is no peg, dollar inflows and outflows do not change reserve money.
For all intents and purposes, reserve money is fixed in the short term. Even if cash from the banking sector increases (say, ATM withdrawals) which is the only way reserve money can change, it may come back the next day. Therefore both the interest rate and reserve money is fixed.
In a floating rate regime the loans that banks can give is automatically limited to the daily loan repayments and net new deposits. As a result, the outflows of dollars have to match the inflows. But in the short term, the exchange rate floats. That is to say, the exchange rate will go up and down based on the timing difference of the inflows and outflows. Outflows will be determined by the credit.
In the meantime, the dollars will remain the net open position (NOP) of banks since the central bank will not buy or sell them. The NOPs themselves will serve as a buffer preventing big changes in the floating exchange rate. NOPs cannot influence the exchange rate on a permanent basis either in a peg or floating rate.
NOP
NOPs are strictly speaking a part of the U.S. monetary base and have nothing to do with Sri Lanka. That is why exchange rate volatility worsens when NOPs are cut. However, if banks overdraw the lender of last resort windows, there could be excess outflows, the currency may weaken and inflation will go up.
It is the job of the central bank not to let this happen and control the reserve money either through a quantity or inflation target. That is why pure floating rates are very strong. When rates are hiked and credit slows, the exchange rate may appreciate.
While NOPs cannot create forex shortages, any NOPs financed by central bank windows or any import financed by central bank credit will create forex shortages. The problem is not with the NOPs or credit but central bank re-financed credit.
SURRENDERS
Exporter and remittance surrenders will also create further pressure on the peg and skew it downwards when the peg is under pressure from excess liquidity (the peg is on the weak side).
This column has explained in the past how the central bank put even more pressure on the currency peg by purchasing dollars coming to the Treasury during currency crises. The negative effect is reduced if the central bank is prepared to re-sell the dollars when import demand comes. If not there is pressure on the currency.
Forex surrenders, like the sales of Treasury dollars to the central bank in previous crises, is one reason that the rupee came under severe pressure and banks had to ration LCs. The surrenders come from the spurious Keynesias belief in a ‘transfer’ problem explained in a previous column.
THE SUPREME PREBISCH IRONY
One of the biggest ironies behind the Latin American central bank debacles is that like East Asia they had the tools to collect large volumes of forex reserve to be used for ‘counter cyclical’ policy.
Raul Prebisch virtually invented the tool to build reserves in excess of reserve money growth, in the form of ‘central bank securities’. In years like 2009, 2013 or 2017, or even the first half of 2019, Sri Lanka built up forex reserves by selling down Treasury bills acquired to create the currency crises of 2008, 2011/2012, 2015/2016 and 2018.
However, a Latin American soft peg styled after Argentina’s BCRA had its own securities to build reserves even if it did not first trigger a currency crisis by acquiring government bonds. Central bank securities (or sell-down of T-bills) are in the nature of a fixed deposit given to people by a bank so that they can keep the book without using the money to consume and import goods.
“To give the central bank scope to influence the money supply through open market operations, it will have the power to buy and sell not only Government securities, but also special negotiable securities which is may itself issue,” an unknown classical analyst wrote in the London based The Banker in July 1950, going through a report done by John Exter, a U.S. money doctor who set up a Latin America style central bank in Sri Lanka abolishing a currency board that kept the country stable through the Great Depression and two World Wars.
Unheeded advice
An unknown classical analyst sounded the warning about Sri Lanka’s Central Bank writing in the London based The Banker in July 1950.
“A device of the same sort was first introduced in Argentina by Dr Raoul Prebisch, in those days before the war when that republic was experimenting in the most heartening fashion with enlightened techniques of monetary management for developing economies.
“The rules then laid down by Dr. Prebisch for Argentina are now incorporated in the Ceylon (central bank) constitution: the bank’s obligations must not be issued for profit and if and when they are later re-purchased, must not be included among its assets but must be immediately cancelled.”
A country would be able to more easily run a ‘current account surplus’ when CB securities are sold to the banking system, by curtailing domestic consumption.
When the securities are repurchased, liquidity is re-injected and forex reserves would fall and the currency could collapse. This is why Singapore, which has a 100 percent reserve-backed currency, based on currency board principles, uses dollar reserves to boost the economy instead of MAS liquidity under rules set by Goh Keng Swee, a classical economist.
Leaving that aside, the supreme irony was that when Prebisch was sacked from the Argentina central bank by the dictator Peron, he had in fact built a considerable arsenal of foreign reserves.
Sri Lanka discontinued the use of central bank securities about a decade ago and did not use them in re-building reserves after the 2012 currency crises after running out of reserves. In that way Sri Lanka’s central bank lost the only good tool that it had and several East Asian monetary authorities (MAS paper, Bank Negara paper and the Exchange Fund Bills of the Hong Kong Monetary Authority) used to good effect to build reserves.
Mercantilists called this phenomenon the ‘Asian Savings Glut’. In GCC countries where it was achieved by selling certificates of deposit, Mercantilists called the phenomenon ‘Petro dollars.’ The UAE Monetary Authority replaced CDs with ‘Monetary Bills’ in 2020.
Any country, including GCC countries, can destroy themselves like Sri Lanka or generate considerable instability if their monetary authorities become more activist and try to boost domestic credit as happened with the Bank Negara Malaysia.
There are two key reasons why any Mercantilist who went to a Keynesian university finds this concept difficult to grasp. Mercantilists believe that a currency falls due to ‘overvaluation’ and some monetary God or monetary overvaluation computer tracks the inflation differential of trading partners, and ‘Hey Presto’ busts the currency when it goes above some number above 100 percent.
While the ‘over-valuation computer’ works in soft-pegs it mysteriously fails to work in East Asia and GCC countries despite the REER index going above 130. But Mercantilists do not worry about such uncomfortable and jarring real life phenomena. They do not worry about the uncomfortable fact that Iran with so much oil has no ‘petro dollars’
In soft pegs like Sri Lanka the rupee collapses despite the REER being below 100. They will continue to believe in the holy truths taught by their Keynesian or Post-Keynesian university professors and throw logic and reason out of the window.
Keynes taught at Cambridge University. Alvin Hansen was at Harvard. They had the tools and the power to mislead a generation of economists – the lost generation – for which countries like Sri Lanka are paying the price to this day.
Sri Lanka’s forex problems are created by the central bank and the Treasury through the control of the 12-month bill yield, the policy rate and overdrawing state banks through central bank window money
THE BUDGETARY PROBLEM VS THE TRANSFER PROBLEM
That is the key reason Sri Lanka printed money from 2015 to 2019 and tried to target output by removing all restraints built by A S Jayewardene, who went to LSE. Policymakers today chase after dollars and try to get loans from China and swaps, in a classic Weimar Republic Young plan style activity and have imposed surrender requirements.
Keynesians in Sri Lanka cannot grasp the reality that remittances and exports are not ‘owned’ by the state and in order to repay loans the state has to get hold of the underlying rupees through taxes or borrowing.
In the 1920s Keynes also believed that there was a ‘transfer problem’ in relation to repaying foreign debt, in addition to the above ‘budgetary problem’. He could not grasp the concept that as soon as the ‘budgetary problem’ was solved in Germany, the ‘transfer problem’ of getting foreign exchange was automatically solved.
If the government wants to get the surrendered dollars (or any other dollars for that matter), it first has to get hold of rupees. This is why Treasury bills have to be sold for real money and domestic private credit crowded out in order to generate dollars to repay debt.
DOMESTIC SOLVENCY
Sri Lanka’s forex problems are created by the central bank and the Treasury through the control of the 12-month bill yield, the policy rate, and overdrawing state banks through central bank window money.
A collapse in tourism revenues cannot cause an exchange rate or forex problem. When tourism sector workers and businesses do not get revenues, their spending and imports will fall in proportion.
If exports fall, imports will fall. If exports rise, imports will also rise, unless the government taxes or borrows from the export sector workers by selling Treasury bills to them or the banks in which they deposit their salaries.
If the central bank sells down its Treasury bills (or sells Raoul Prebisch sterilization securities) outflows will be curtailed and it can buy dollars in the market to rebuild reserves and give to the government to repay loans.
Stopping imports selectively will not serve any purpose, since credit and printed money will go to other areas where credit is permitted, as happened when the economy recovered in the first quarter. Imports bounced back to 1.7 billion dollars a month, partly driven by printed money.
The reason this country got into so much trouble after the end of a war is due to blindly following a false ideology in both monetary and fiscal policy. Latin American countries also get into trouble without war due to the domestic operations of the central bank.
Sri Lanka’s crisis can only escalate as long as there are ceiling prices on Treasury bills, which are bought by the CB to create Treasury bills and as long as there are exporters and remittances surrenders whose liquidity is not mopped up.
THE REQUIRED RATE
To stop liquidity injections, rates have to be raised. Surrenders are not needed. Raising oil prices was a good move but not enough.
If policy makers want to minimize the required rate to rebalance the economy, value added tax has to be raised and the state spending bill has to be trimmed. Growth is not going to solve the problem, since Latin America style growth is going to tip the balance of payments over.
An IMF program will allow debt repayments to be delayed also allowing lower interest rates than otherwise. Any kind of state austerity will reduce the equilibrium interest rate needed to balance the BOP. If not the state austerity – cutting of wages and benefits of state workers – will come through depreciation only.
Sri Lanka is not Greece as this column has said before. In Greece the anti-state austerity brigade could not destroy the currency and impose austerity on the aged, the pensioners and the sick through depreciation and Greece was forced to cut spending through ‘state austerity’ and also imposed losses on bond holders.
But the poor pay the price in Latin America repeatedly and they will pay the price in Sri Lanka. The widening parallel exchange rate is a sign of things to come. It now seems almost too late. More downgrades are around the corner. Rating agencies do not really understand soft-pegs but they understand reserve losses.
Downgrades are around the corner. But it is never too late to take action. The central bank policy errors in 2015 and 2018 were fixed with 2-3 percent growth.
After 2019, the non-credible peg errors have been compounded with tax cuts and a sovereign confidence shock and default which Sri Lanka did not have to battle with in earlier soft-peg crises.
This time, the crisis may have to be fixed after a ‘sudden stop’ of external financing. Sudden stop events lead to very sharp economic contractions. So do very steep depreciations. It is too late to set up a currency board, and too late for official dollarization at these levels.
Sometimes investors are quite prepared to take losses. Some of Sri Lanka’s bonds are already held by vulture style funds who have bought at low prices. They can take a hit.
Like rating agencies they do not really understand soft-pegs as commentaries on Sri Lanka bonds by various investment banks show, but they know forex reserves losses. IMF backed Latin America debt restructuring is also something they are familiar with.