Should foreign reserves finance the public budget?

What should one look at to assess the possibility of using official foreign exchange reserves for public investment projects? In a world of fungible resources, this essentially means using reserves to fund the government budget. A key judgment call is the adequacy of reserves. Does Bangladesh have excess foreign reserves which could be better used in public infrastructure projects instead of borrowing money from international financial institutions, markets or bilateral partners?

A confidence builder   

Prudent reserves, along with sound policies, bring significant benefits. They reduce the likelihood of balance-of-payments crises, help buffer pressures on exchange rates due to disorderly market conditions and provide space for policy maneuvering. It is a key indicator of sovereign risk--the risk that the government or an entity guaranteed by the government defaults on its debt. The Economic Intelligence Unit in its June 2020 Country Risk Update on Bangladesh says: 

"The government will record a large fiscal deficit in fiscal year 2020/21…because of the need for fiscal support owing to disruptions caused by the coronavirus…pandemic. However, high levels of foreign exchange reserves and low levels of public debt to GDP compared with other B-rated countries will allay concerns over the country's ability to repay its debt."

Strong reserve level gives foreign investors and credit rating agencies added comfort that the government can meet its debt obligations despite a deteriorating fiscal outlook.  Since reserves are invested in liquid assets, they earn a low rate of return than if they were used in longer-term illiquid assets. However, the purpose of holding reserve is to mitigate risks, not maximise returns. A realistic approach is to identify the key contingencies, ascertain the minimum reserve needed in each of these contingencies and choose the maximum from these minimums as the "adequate" level of reserves.

External vulnerability increasing  

A key indicator is the external current account deficit which increased from $765 million at the end of September 2019 to $4.4 billion at the end of May 2020.  This was driven by a sharp fall in exports, even before Covid-19, and rapid deceleration of remittance growth since February. Bangladesh's trade gap has widened despite declining imports. How a narrow export base can haunt is most evident from Bangladesh's experience in FY20.  

A large shock to garments, total exports are down by nearly 18%. Making matters worse, except for jute goods and tea, all other non-garment exports also recorded large declines. The uptick in remittances in June was predominantly due to transitory factors such as transfer of savings by returnee migrants.  The slide in the current account deficit will most likely continue into the near future. The increasing current account deficit did not create a financing problem because of budget support from the IMF, ADB, WB and so on, disbursed in the second half of FY20.  Disbursement of multilateral loans increased by $1.6 billion during March-May. These are one-off receipts that cannot be expected to sustain.

Bangladesh Bank (BB) has revealed a preference for stable exchange rates. International reserves allow the BB to make the implicit guarantee that it will be able to bridge the gap between demand and supply to keep the exchange rate fluctuation within a tolerable band.  Imagine what would happen if BB can no longer fulfil this unwritten assurance.  Excess exchange rate volatility will make everyone nervous.  Instability in the foreign exchange market can have wide-reaching consequences. Costs of imports become uncertain. Both local and foreign investors care whether the current exchange rate is sustainable.  It affects the risk of exchange rate-related losses if they were to invest in the country.

Adequacy can be elusive

The existence of adequate reserves does not, by itself, eliminate the risk of market pressure.  The notion of reserve adequacy is not static. It is associated with the likelihood of financial crisis and exogenous shocks. When the fiscal stance is expansionary, the minimum level of reserves needed for market confidence rises.  Add radical uncertainties such as the vagaries of the economic impact of the worst and still very poorly understood virus such as the coronavirus, the adequacy level rises even more. 

Reserves provide the liquidity for crisis prevention and mitigation. While other instruments, such as official credit lines, are also external buffers, they principally act as a complement to the official reserves. Reserves provide the space to respond to shocks. This space diminishes as fundamentals such as exports, remittances and fiscal deficits deteriorate. The crisis created by the virus all over the world has heightened the precautionary demand for reserves. 

There is little consensus on what constitutes an adequate level of reserves from a precautionary perspective.  Metrics traditionally used are narrowly based and sometimes provide conflicting signals. Newer approaches tend to be hostage to stylized modeling assumptions and calibrations. Under the circumstances prevailing currently, ascertaining the adequacy of reserves requires stealing a page from chaos theory. It is important to be aware of the undeniable fact that the rhythm of the economy is not predictable like the sun rises in the east. There are many uncontrollable forces around us that can make the other shoe fall at any moment. A tiny invisible virus in Wuhan scattered in all directions around the world in quick time creating the deepest economic crisis, arguably, since the Great Depression. Science fiction has turned into a depressing science reality.

How much is enough?

Economists commonly use, in normal times, two rules of thumb as starting points in answering this question.  Reserves should cover three months of projected imports.  There is no scientific basis for this ratio.  It is a commonly accepted threshold blessed by the IMF.  It focuses solely on the flow of goods and services. In a country like Bangladesh, this is an important shortcoming.  Importing more goods and services than we export, it is necessary to attract more capital than we launder out of the country.

A second widely used indicator is the ratio of reserves to broad money.  This is the amount of printed currency outside the banking system as well as checking and time deposits. Banks never have all the deposits in their vaults because only a fraction of the depositors demand access to their funds on any given day.  The same is true for reserves. Only a fraction of the holders of local currency would want to convert into foreign currency on any given day.  The benchmark considered adequate in this instance is that reserves should be around 20% of broad money to cover against money supply being converted into foreign currency in the short to medium-term for purchasing goods and services or "investments" abroad.

These two ratios are useful but not sacrosanct. They ignore factors that need to be considered when determining the benchmark. The challenge is to be sure that the thresholds are adequate and give the captains of macroeconomic management enough time to prepare. Following conventional rules can fail to signal distress before it is too late.

Downside risks loom large  

Bangladesh's current level of reserves appears more than adequate using conventional rules. The $36 billion reserve is equivalent to about 7.1 months of estimated merchandise and services imports in FY20 and 21% of broad money. However, the reality is more complicated.  External risks related to the narrow exports base and anti-migrant sentiments as well as policies in countries hosting Bangladeshi labour remain high. Bangladesh is vulnerable to natural disasters requiring larger and quickly available foreign currency to finance recovery efforts. Foreign currency flows quite freely in and out of the country despite a complicated set of controls managed by BB.  

All these imply the need for a larger cushion than prescribed by the conventional metrics. In the macroeconomic framework recently agreed with the IMF, the reserves were projected at 5.6 months of imports by the end of FY20, equivalent to $29.5 billion.  This may suggest excess reserve of $6.5 billion compared to the current level.  But wait.  The overall balance of payment deficit is projected to rise from $3.5 billion in FY20 to $4.8 billion in FY21. Avoiding any reserve drawdown requires having $34.3 billion, leaving just $1.7 billion as excess. Even this may not be safe, considering the difficulty of achieving the projected 7% remittance growth from a historic high of over $18.2 billion in FY20.  The expenditure levels will increase while revenue growth is destined to be well below historic averages. The fiscal deficit will most likely overshoot the 6% of GDP target.  A significant part of the deficit is likely to be monetised.

Better safe than sorry

Since the Asian Financial Crisis, a growing number of countries have maintained vast stock of reserves. This was exacerbated in the aftermath of the Global Financial Crisis, rising to as high as 30% of GDP.  Take India. Their level of reserves covers 13 months of imports, equivalent to 20% of GDP.  Reserves in months of imports of goods and services was 7.6 in Cambodia and Indonesia, 7.9 in Philippines and 6.3 in Indonesia at the end of 2019.   

In deciding how much, if any, reserves can be used to finance public investment projects, it is critical to assess rigorously and independently what foreign exchange reserve level is the minimum needed to comfortably meet the foreign exchange obligations. The failings of governance in the government expenditure management framework is also a key consideration. There is not much comfort in the existing macroeconomic framework to warrant voluntary parting with reserves as an alternative to seeking external financing.