It may the first time it has been tried in India, but 'raiding' the central bank's balance sheet and financial resources to meet government spending needs in not uncommon. In 2015, for instance, the US House of Representatives reduced the US Federal Reserve's (the Fed) capital surplus (balance sheet reserves, in other words) from $30 billion to $10 billion to finance highways.
"Financing federal fiscal spending by tapping the resources of the Federal Reserve sets a bad precedent¦." said Janet Yellen, then Federal Board chairperson, while testifying before the US Congress, adding, "It weakens fiscal discipline."
The US Congress had a Republican majority in both the House and the US Senate.
But the Republicans didn't stop there. In February 2018, they took out another $2.45 billion, reducing the reserve to $7.5 billion. This time, the money was used to prevent a government shutdown because the Congress couldn't reach agreement on the budget for the year. This, after the Fed had already transferred $80 billion to the US Treasury as 'dividend' in 2017 (the Fed answers to the US Congress).
It's not just the US. In 2012, Argentine President Cristina Fernandez sent a bill to the legislature that would allow her to take money out of the central bank's reserves. In 2013, Australia's treasurer Wayne Swan wrote to Glenn Stevens, then governor of the Reserve Bank of Australia, demanding that the central bank pay A$500 million of its earnings to the Treasury as dividend (Swan had said that he would deliver a surplus Budget in 2012-13).
Other countries have tried or considered it too: Ireland, Azerbaijan, Greece, Hungary, Zimbabwe, to name a few. Many others have considered using foreign exchange reserves to fund infrastructure. A decade ago, the then Planning Commission vice-chairman Montek Singh Ahluwalia made a serious case for it. Thankfully, nothing happened.
Now, finance minister Arun Jaitley wants to take Rs 3.6 lakh crore out of the RBI's reserves. That's roughly 2 percent of GDP, not a small number. The rupee equivalent of the US case cited above would be Rs 1.65 lakh crore ($22 billion), one-tenth of 1 percent of US GDP ($19.5 trillion). This is a terrible idea, as the following paragraphs will elaborate.
It's not just about the size or the amount: it's really about what such a step will do the balance sheet of the RBI and the wider impact that can have on the institution's financial resources, and by extension its credibility in financial markets (think about the impact on capital outflows!). To get a better handle on that, let's take a closer look at the structure of the RBI's balance sheet.
Structural analysis of the RBI's balance sheet
The RBI's liabilities fall into four categories: 'equity' capital, deposits, provisions and currency issued (most of which is in circulation). On the asset side, there is gold and bullion (divided into roughly two equal buckets), investments (foreign and domestic) and loans to central and state governments.
The RBI balance sheet is divided into two parts: currency-related and banking-related. Currency liabilities are backed by one bucket of gold bullion plus net foreign assets (NFA): mostly foreign currency investments made out of our foreign exchange reserves.
The paid-up 'equity' capital of the RBI is Rs 5 crore, and a reserve fund of Rs 65 crore; both these haven't changed in decades. The RBI is also a bank, so carries deposits and makes loans: to central and state governments, and mostly banks for clearing and settlement purposes and cash reserve requirements or maintaining CRR, in other words. International financial institutions and foreign central banks, also have deposits, but meagre amounts.
'Other provisions and liabilities' are backed by banking-related assets. Investments (domestic and foreign), gold and bullion (the second bucket) and loans to governments make up almost all these assets. It's the 'other provisions' that the government is after, and that which is the focus of the current drama. At end-June 30 2018 " the RBI's financial year is July-June " this account head was Rs 10.463 lakh crore, up by 30 percent from the previous year's Rs 8.95 lakh crore. It is from this account that the government wants to extract Rs. 3.6 lakh crore.
Financial buffers; why we need them
There are two big components to these provisions: a contingency fund, or CF (Rs 2.32 lakh crore, and a currency and gold revaluation account, or CGRA (Rs 6.92 lakh crore). The contingency fund is a specific provision for risks of exchange rate depreciation, or losses from open market operations (OMO) and changes in the market value of the RBI's domestic investments. The CF has been 'flat' between 2014 and now.
The CGRA was created to account for potential market risks from exchange rate appreciation or depreciation, shifts in interest rate risk and changes in gold prices. Central banks take two approaches to addressing market risks on their balance sheets. The equity accounting approach requires each specific risk be accounted for and provided for separately before arriving at net income; one consequence is that the central bank can have negative equity or net worth if losses on these risks exceed income.
The liabilities approach " which the RBI, the European Central Bank and many Eurosystem central banks take " creates financial buffers for managing general risks and revaluing unrealised gains on both domestic and foreign investments. This way, the RBI removes the risk of prematurely including non-cash items in the dividend it pays to the government.
The operative term here is 'unrealised' gains or losses. This is essentially a revaluation reserve, not a 'cash' reserve. It is also volatile, because of daily changes in exchange rates and interest rates. Over the last five years, it has varied from Rs 5.72 lakh crore in 2014, through Rs 5.59 lakh crore (2015), Rs 6.37 lakh crore (2016) and Rs 5.3 lakh crore (2017) to Rs 6.91 lakh crore in 2018. The 30 percent increase from 2017 is mainly on account of rupee depreciation and higher gold prices.
What does an analysis of income and expenditure tell us? The RBI made Rs 50,900 crore in interest and other domestic income and Rs 27,400 crore from foreign sources in 2018. Total expenditure (including a transfer Rs 14,100 crore to the CF) was Rs 28 300 crore. The profit of Rs 50,000 crore was transferred in its entirety to the government.
In fact, every year since 2014, the RBI has transferred all the profits it has made to the government; as noted above, the CF has not varied much from Rs 2.21 lakh crore in 2014 to Rs 2.32 lakh crore in 2018. At the same time, the CF as a percentage of total assets has declined from 9.3 percent in 2013 to 6.4 percent in 2018. Internal working groups within the RBI had concluded that the appropriate level of the CF was 12 percent of total assets. If we are not adding to the buffers, should we be reducing them further?
As long as a central bank can print money, it can create seigniorage " the difference between the cost of printing currency and its face value " and that means earnings. Experts point out that a central bank can thus never be 'technically insolvent' as long as it has the ability to print money, even it has 'negative equity'. But using that as a basis for drawing on the CGRA is not a great idea. Here's why.
Central banks do not have stable or voluminous sources of earnings. Between 1984 and 2005, there were 43 cases of central banks having made losses in at least one year (out of 108 central banks). Between 1987 and 2005, 15 central banks in Central and South America made losses for five years running; 8 of them made losses for more than 10 years in a row. So central banks by nature are not profitable.
So what are the implications? Alain Ize, a researcher at the International Monetary Fund (IMF), looked at the balance sheets of 87 central banks with positive and negative structural profits. His study found that in 2003, the average inflation of those countries whose central banks had positive profits was one-third of countries in which central banks had negative profits - 3.5 percent versus 9.5 percent (Yes, the RBI was included in that sample).
Other researchers using different years and different samples of central banks have broadly confirmed those findings in 2011. The message is simple: Central banks with weaker finances tend to have higher inflation outcomes " twice as high at least. Given that one of the principal objectives of a central bank is inflation control, a financially weakened central bank would be hard-pressed to fulfil that objective. Experts call it 'policy insolvency'.
The transfer of Rs 3.6 lakh crore to the government coffers is a gigantic dividend: What economists call asymmetric distribution.
In 1998, the Bank of Israel made significant gains on exchange rate depreciation. Given its accounting policies, the Bank of Israel had to transfer all the gains " NIS 9 billion or Israeli new shekels " to the Treasury in February 1999, even though by then the gains had been reversed! The next year, the Bank of Israel ended up with negative equity " to the tune of about NIS 9 billion.
Central banks use accounting policies to avoid such distribution asymmetries; the RBI does not recognise revaluation gains as income, preferring instead to take the liabilities approach described above. After the 1998-99 debacle, the Bank of Israel changed its accounting policies; unrealised gains from exchange rate depreciation are no longer treated as income, but put into a revaluation account like the CGRA.
Forcing the RBI to pay an asymmetric dividend would effectively hinder the central bank from fulfilling its primary purpose: Inflation control. This is not about central bank independence or the nature of the relationship between the RBI and the finance ministry. This is about simple, sensible economic policy. Good economics is also good politics.
(The writer is a former senior journalist and communications consultant)