Making the forex reserve policy rule work

The government has decided to use part of the official foreign exchange reserves for investment in infrastructure. Henceforth, the Bangladesh Bank will hold 6 months' equivalent of imports in liquid assets and make a maximum of $2 billion available annually for the Bangladesh Infrastructure Development Fund (BIDF).

The Bangladesh Bank will lend to a financial intermediary, which in turn will lend to state-owned enterprises (SOEs) to build infrastructure. The sectoral and institutional reach of the BIDF may be expanded as time goes by depending on how it works.

The decision has instituted a new policy rule on the use of foreign exchange reserves. What must happen to make it work? The answer depends on maintaining adequate reserves, the operationalisation and adherence to the reserve allocation rule and how the foreign currency financing channelled through the BIDF is used.

Why reserve adequacy matters

Official foreign exchange reserves are held for supporting and maintaining confidence in monetary and exchange rate policies. It limits external vulnerability by maintaining foreign currency liquidity to absorb shocks during times of crisis or when external borrowing is difficult.

Several economies have recently experienced significant expansions in their balance of payment surpluses, including India, China, Taiwan, and Vietnam. Their reserve positions have surged. Reserves in India are nearly 18 months' equivalent of imports. All these countries have raised the scale and persistence of foreign exchange intervention to resist the appreciation of their currencies. These actions occurred mostly during a period of dollar weakness.

Investors and rating agencies derive comfort from the adequacy of reserves. Safety and liquidity provide a level of confidence to markets that the country can meet its external obligations and buffer the impact of unforeseen emergencies while limiting exchange rate volatility. The recent increase in food, medicine, and intermediate goods import bill is a good reminder of how import pressures arise when nature strikes.

The Bangladesh Bank uses foreign exchange reserves for the orderly movement of the taka. It sold over $4.6 billion in FY2018-19 to stem depreciation pressure on taka and has bought $6.3 billion this fiscal year to prevent the appreciation of taka. It has always intervened transparently in the market to maintain the stability of the taka against the US dollar. This is the official policy stance. Liquid reserves are handy for selling dollars when the taka weakens and buying when taka strengthens.

Rationale for seeking higher returns

It is important to keep an eye on returns as a secondary objective in reserve management. Bangladesh Bank's income from the foreign currency financial assets decreased to Tk49.7 billion in FY20 compared to Tk59.9 billion in FY19, according to the central bank's Annual Report 2020. These constituted 2.2% and 1.6% of Bangladesh Bank's total foreign currency assets, respectively. Notwithstanding the decline, these returns are not too bad, considering they were from safe assets.

Interest rates in the US and Eurozone have been near zero, with some securities even trading at negative yields, for a while now. Monetary easing by the US Federal Reserve and in other advanced economy central banks has reduced the attractiveness of traditional reserve currencies. Inflation-adjusted rates in the US are still negative despite doubling of yields on five-year US Treasury since the start of 2021. Rising yields currently in the eurozone may not sustain given a weaker economic and inflation outlook and their need to keep borrowing costs lower for longer to secure economic recovery.

If adequate liquidity is maintained, where "adequate" is prudently determined, then anything beyond could be better placed in other assets that lack liquidity but yield higher returns. The BIDF intends to step up infrastructure investment since private investment in infrastructure has been limited, except in power generation, telecom, and the Internet to some extent. Public investment is constrained by abysmally low revenue generation and access to finance from a handful of multilateral institutions and a few bilateral donors. Some of these (IDA credit) are still cheap and others (suppliers' credit) expensive.

Operationalising the policy rule

The current level of reserves, at around $43 billion, is adequate for self-insurance and access to dollar liquidity as the first line of defence. It is equivalent to over 9 months of imports of goods and services; 3.5 times short-term public and private external debt, 67% of total external debt (including public guarantees), and 24% of broad money.

Operational judgment on "adequacy" required to ascertain the size of spendable "excess" reserves depends on the outlook. Adequacy for the purpose of financing the BIDF is linked to the months of imports metric. Assessing how much more than 6 months adequate requires a choice between backward- and forward-looking assessments of both the numerator and the denominator.

Assuming the numerator (reserves) will continue its recent rising path may not be prudent. Certainly, using the monthly import (denominator) levels seen during the pandemic can be very misleading at a time when import demand is poised for a strong recovery and import prices could rise.

Bangladesh Bank's foreign exchange reserves increased by over $10 billion since the end-June 2018. Much of these came from a decline in imports, a surge in remittances, and increased financial account inflows (trade credit and foreign assistance).

Goods and services imports amounted to $5 billion per month in FY20 compared with $5.5 billion in FY2018-19. Remittances increased from $1.3 billion per month in FY19 to $1.5 billion in FY20 and further to $2.2 billion in the first seven months of FY20. Trade credit turned from $3.5 billion outflows in FY19 to $1.6 billion inflow in July-January this fiscal year. Net foreign assistance increased from $425 million per month in FY19 to $475 per month in FY20.

The durability of the increase in reserves cannot be taken for granted. The current account surplus dropped from $4.3 billion in July-December to $2.2 billion in July-January, a decline of more than 45% in a month. Soaring prices of consumer items, soybean oil, fuel, and intermediate inputs in the international market; food production shortfall; and domestic demand recovery led to a $6.7 billion increase in import bills in January. Remittance growth slowed from 48.3% in July-September to 30.7% in July-February. Net foreign assistance declined to $347 million per month in July-January.

If domestic demand turns around further in the latter part of this year and as international commodity prices increase, monthly import bills for goods and services could rise to $6, or even $7, billion in FY2022-23. This implies adequacy level ranging between $36 to $42 billion going forward. A simplified judgment call could be the arithmetic average of the two, $39 billion, as "adequate" reserves.

The policy rule is to make a maximum of $2 billion available each year from the difference between "actual" and "adequate" reserves. Actual reserves change frequently. Reserves peaked at over $44 billion on 24 February before falling back to $42.6 billion on 15 March. A recent low was $39.3 billion last September. Which value to pick?

One could use, for instance, the average of last quarter or last two quarters. The September-February average, for instance, is $41.6 billion. This results in a spendable excess of $2.6 billion, safely above the $2 billion cap. The Bangladesh Bank needs to specify these operational details for ensuring the adequacy of reserves.

Both the numerator and the denominator of the metric can change in either direction going forward. The rule itself will need to be periodically reviewed to account for structural and cyclical shocks to external vulnerability. The "6 months" cushion is prudent under current circumstances, but that could easily change.

Managing the inflation risk

Recent forex purchases by the Bangladesh Bank created large excess liquidity in the banking system. Excess liquidity also resulted from lowering of taka liquidity ratios, direct liquidity support, and initiation and expansion of various refinancing schemes. Excess liquidity has persisted despite significantly reduced lending rates. There are therefore some pre-existing worries about kindling consumer and asset price inflation as banks push for better volume to boost their revenues.

The use of the BIDF could exacerbate the inflation risk, depending on how the recipients spend it. If all of it is spent on goods and services sourced from abroad, there is no additional domestic demand and therefore no "inflationary" effect. The country will lose reserves if all of it is used for imports, leaving the exchange rate unchanged. Increased demand from the infrastructure project will be matched by supply of foreign currency from the reserves.

Inflation could result if the loan from reserves is used to finance local costs. As local workers, suppliers and contractors get paid in taka, they add to demand, which then multiplies. This may be further magnified by Bangladesh Bank's intervention needed to maintain exchange rate stability. Reserve accumulation occurred because private parties sold dollars for taka at the existing exchange rate. As the borrower cashes out to pay local bills, these official dollar reserves will be re-cycled back into the foreign exchange market. The Bangladesh Bank must then repurchase the dollars to prevent appreciation of the taka. The incidental effect of this "round tripping" would further expand the monetary base.

The inflation risk is a serious consideration especially when the BIDF is scaled up beyond its initial mandate to finance projects in public and private sectors with greater local cost financing.

Handle with caution

The returns from infrastructure investments may not be appropriable as intended. The appropriation risk arises fundamentally from mismanagement and a symbiosis between private rent-seeking and public corruption. Infrastructure projects in Bangladesh yield low returns due to time and cost overruns. Returns are not inherently low in many cases, but they wither into the pockets of rent-seeking contractors and corrupt public officials. Inefficiencies produce losses for all while the rent seeking-corruption nexus redistributes towards the connected interest groups at the expense of the intended beneficiaries.

There could be moral hazards and herd effects. The BIDF will lend through state-owned banks to the SOEs. Both are problematic enough on their own. The loans are backed by a sovereign guarantee promising to discharge the liability in case of default by SOEs. The guarantee could breed profligacy and shirking. The prospect of SOEs to earn in foreign currency from infrastructure is certainly uncertain. Having set a precedent, the BIDF risks opening a pandora's box for less desirable uses of forex reserves in future. Several business groups have been seeking loans from reserves.

Fiscal prudence and transparency require not allowing extra-budgetary actions to proliferate. The "self-reliance" argument works both ways. Self-reliance for infrastructure finance must not be at the expense of self-reliance for managing balance of payments and exchange rate vulnerability.

Let us not forget that strong forex reserves allow a country to negotiate external finance from a position of strength. Weak reserves indulge no such luxury.