This article is more than 2 month old.

RBI dividend to government: Justified for now, but beware of setting a precedent

Mini

The RBI gave an estimate of Rs 45,000-50,000 crore for dividend in December. How did it manage to account for double the amount of dividend?

At the outset, RBI’s extra-large dividend of Rs 99,122 crore needs to be welcomed. The nation is in trouble. A large number of citizens are sick. Economic activity is minimal. The government may fall short of tax and non-tax receipts, even as health expenses mount and hence money printing by the central bank is perfectly justified. If not now, then when?
That said, we need to understand the scale of the dividend and its implications, and the way in which the RBI appears to have generated this surplus. First up this latest dividend is for the 9 months-ended March 31 (this year the RBI is changing its financial year to start from April; until last year RBI’s year began on July 1). So RBI has paid out Rs 99,122 crore for 9 months as against Rs 57,000 crore it paid for 12 months last year. Also, this Rs 99,122 crore is almost twice the amount the budget estimated from RBI. The BE for dividend from RBI and PSU banks for the current year was estimated at Rs 53,510 crore.
If on February 1, the government estimated that only Rs 53,510 crore will come as dividend from RBI and PSU banks put together, then it is likely that in December, when asked to estimate its likely dividend by the finance ministry, the RBI gave an estimate of Rs 45,000-50,000 crore. Then how did it manage to account for double the amount of dividend?.
The RBI makes money through what is called seigniorage. This means when it prints rupees to buy government bonds, the cost of printing is next to nothing. So the entire incremental bond portfolio becomes its income. Besides this, RBI makes money on the foreign bonds and deposits in which its invests its reserves.
The RBI's expenses include about 12,000 crore of establishment expenses and what it pays banks in the reverse repo window, which was rather large from July 2020-March 2021. But its biggest expense is provisions. The Jalan committee requires RBI to maintain equity (or retained earnings) at 5.5-6.5 percent of its total assets. It also requires RBI to maintain total economic capital (retained earnings plus revaluation reserves) at 20.8-25.4 percent of total assets.
Counting all this market expectation was that RBI would pay a dividend of Rs 60,000-65,000 crore. Prof Ananth Narayan, writing for the Observatory group, estimated the dividend at Rs 60,000 crore and Barclays economist Rahul Bajoria estimated the dividend at Rs 65,000 crore. Just to repeat, this is a 9-month dividend and compares with Rs 57,128 crore last year for 12 months.
How did RBI manage to pay at least Rs 40,000 crore more than market estimates? It is likely that the RBI has used the tiny crack provided by the Jalan committee.
Let me explain: Before the Jalan committee recommendations, the RBI derived very little profit from its dollar sales. This is because every year RBI would revalue all its reserves at the current exchange rate and transfer the gain/loss to the revaluation reserve. Dividend was not paid out of reserves, but from the profit and loss statement. The Jalan committee, in Aug 2019, said RBI, when it sells dollars, may consider the historical average price at which it had acquired dollars as the purchase price, deduct that from the price at which it sold dollars in the current year and give the difference as dividend to the government. While this formula can’t be faulted, the crack in the door is, if RBI wants to declare a higher dividend, it can do a whole lot of sell-buy swaps, and in the process, “account” for a higher dividend to the government.
It is possible the RBI has done exactly this in January, February and March to arrive at a large dividend. Here are the total sales and purchases of dollars done by the RBI in the past 3 years (note the jump in 20-21):
FY19FY20FY21
Purchase ($ bn)40.872.2140.5
Sale ($ bn)56.127.185.24
Source: RBI Bulletins
*calculated by CNBC-TV18 & Prof Ananth Narayan in the Observatory from monthly data given in RBI Bulletins
As Table 1 shows, in 20-21, RBI has bought at least twice the amount of dollars it did in 2019-20; and sold over 3 times what it did in 2019-20. Now here’s the month wise break-up of sale and purchase of dollars in 2020-21:
Purchase ($ bn)Sale ($ bn)
July16.90.93
August8.523.21
September13.325.17
October15.640
November14.294.02
December10.016.02
January18.2315.37
February23.3524.57
March20.2525.95
Total140.585.24
Source:  RBI Bulletins
As is clear from the above table, 75 percent of the dollars sold in FY21, were sold in the last three months, and likewise about 45 percent of the dollars purchased in FY21, were bought in the last three months. The justification for such large gross sale and purchase of dollars in the last three months is not very clear, when the net action each month was much smaller. One can’t recall any obvious large mismatches of inflows/ outflows in the market during those months.
It is tempting therefore to conclude that faced with a demand for a bigger dividend, RBI used this route of selling spot dollars and buying forward, merely to “book” profit on dollar sales and thus account for a higher dividend. Many bank and corporate treasuries do resort to selling in-the-money securities at the end of the year to “show” profit. But a central bank, in charge of the nation’s exchange rate, resorting to these gimmicks, can be dangerous.
Once the FX market smells that RBI is under pressure to generate profit by selling spot dollars and buying forward, it can push up forward premiums. Indeed that already seems to be the case with forward premiums at historic highs in the past few months.
And then there are other worries from this high dividend, What if the government starts expecting a one lakh crore dividend from RBI as the new normal, knowing as it does, that RBI can do sell-buy swaps and book profits.
All RBI dividend is a matter of printing money. If a central bank prints much more money in a year than the increase in the amount of goods and services, it will lead to inflation. Now in a pandemic year, the RBI may be justified in excessive money printing because given the steep fall in demand ( for goods and for loans), excessive money supply may not lead to inflation. If RBI wants to help in a pandemic, it could well have asked the centre and states to issue vaccine bonds and merely printed money to buy them. Was there a need to use a devious means of selling dollars and buying back in forwards merely to generate profit?
More generally, the global economy may be heading towards inflation and this may overflow into India. Under such circumstances, RBI monetising a large deficit, year after year, is a recipe for inflation; FX markets are fickle creatures.
India was a picture of resilience in 2011 and 2012, and became one of the fragile 5 in a year, in 2013. Short point, the RBI is on a slippery slope with regard to the size of the dividend and the manner in which it has generated this “surplus”. While it is justified in a pandemic, one hopes the macros don’t turn negative anytime soon, so much so, that we live to regret this monetisation.

Market Movers

CompanyPriceChng%Chng
Kotak Mahindra1,652.85 -45.10 -2.66
Cipla889.90 -23.20 -2.54
Tata Motors284.45 -6.95 -2.39
Dr Reddys Labs4,731.75 -111.60 -2.30
M&M728.70 -16.85 -2.26
CompanyPriceChng%Chng
Kotak Mahindra1,653.20 -44.75 -2.64
Dr Reddys Labs4,730.20 -114.15 -2.36
M&M728.45 -16.95 -2.27
Power Grid Corp228.05 -4.05 -1.74
NTPC116.75 -2.00 -1.68
CompanyPriceChng%Chng
Kotak Mahindra1,652.85 -45.10 -2.66
Cipla889.90 -23.20 -2.54
Tata Motors284.45 -6.95 -2.39
Dr Reddys Labs4,731.75 -111.60 -2.30
M&M728.70 -16.85 -2.26
CompanyPriceChng%Chng
Kotak Mahindra1,653.20 -44.75 -2.64
Dr Reddys Labs4,730.20 -114.15 -2.36
M&M728.45 -16.95 -2.27
Power Grid Corp228.05 -4.05 -1.74
NTPC116.75 -2.00 -1.68

Currency

CompanyPriceChng%Chng
Dollar-Rupee74.3750-0.0850-0.11
Euro-Rupee87.9670-0.0450-0.05
Pound-Rupee103.39300.03400.03
Rupee-100 Yen0.6755-0.0027-0.40