Raghuram Rajan: The independence of the central bank

Remarks by Dr Raghuram Rajan, Governor of the Reserve Bank of India, at St. Stephen's College, New Delhi, 3 September 2016.

The views expressed in this speech are those of the speaker and not the view of the BIS.

Central bank speech  | 
05 September 2016
PDF version
 |  6 pages

Good afternoon. It is great to be invited to speak at St. Stephen's College. In 1980, I toyed with the idea of joining my best friends in applying for admission to the BA in Economics here. Because I had worked so hard for the IIT exam, however, I succumbed to the sunk cost fallacy and studied Electrical Engineering. I don't regret a moment of that misspent youth but I hope you will grant me temporary membership of your club today!

Over the last few weeks, I have outlined the RBI's approach to inflation, distressed debt, financial inclusion, banking sector reform, and market reform. Today, I'd like to first discuss why central banking is not as easy as it appears (just raise or cut interest rates!) and why it needs decisions, sometimes unpopular or hard-to-explain ones, to be made under conditions of extreme uncertainty. This will then lead in to my arguments about why we need an independent central bank.

To set the stage, think back to the summer of 2013. After Chairman Bernanke hinted that the United States might end quantitative easing, capital started flowing out of emerging markets, especially those deemed to have weak macroeconomic fundamentals. India, with high fiscal deficits, high current account deficits, and near double digit inflation, was deemed one of the "Fragile Five" countries. The rupee plunged as debt investors scurried to safe havens. The talk in international markets was that the rupee could reach a hundred to the dollar, and some international commentators, knowing little about India, talked about an Indian "crisis".

In early August 2013, I was appointed to succeed Dr. Subbarao as Governor of the Reserve Bank of India. From the Finance Ministry, I went to the RBI to prepare for the transition.

We had to get back the confidence of international investors. And that meant convincing them that India was still an attractive place to invest in, despite the then paralyzed economic reforms. Furthermore, they had to believe that the rupee would retain its value going forward.

So I went from room to room asking RBI staff what they were working on, and what ideas for financial sector reform they were prepared to contemplate. We packaged our ideas into a reform program over the medium term that could change the narrative on India, especially given the fiscal reforms the government had initiated.

To bolster the value of the rupee, we had to give investors more certainty that future inflation would be low. After all, it was primarily India's higher inflation with respect to the rest of the world that led to periodic abrupt falls in rupee value. Then Deputy Governor and now my successor, Dr. Urjit Patel, agreed to prepare a report on how we could move to a new inflation-focused monetary regime. That would help very much in the medium term, but in the short run we had to establish that we were not heading towards crisis. The easiest way to demonstrate that was to show we could raise plenty of foreign exchange.

My thought turned to an iffy scheme that had been presented to me in North Block. Essentially, bankers had told us they would bring plenty of dollars in as FCNR(B) 3 year deposits, which they would convert to rupees and invest in India. In return, they wanted a cheap rate at which they could convert rupees back into dollars 3 years hence. So long as the Reserve Bank could be trusted to provide the forward dollars, this was a great deal for the banks - they got rupee interest income and a guaranteed cheap price at which they could swap maturing rupees back into dollars.

As I have said elsewhere, I did not think much of this scheme when I first heard of it, dismissing it as a clever ploy by bankers to get a subsidy from a country in trouble. But it refused to go away, and my old colleagues at the finance ministry thought it worth a try as the crisis of confidence worsened.

I too became more favourable after thinking carefully about it. First, I weighed the balance of risks. If we did not move the rupee back to fundamental value, every 1 rupee rise in the dollar-rupee rate would costs us Rs 40,000 crores more in import costs. Assuming the rupee was undervalued by Rs 3 for a couple of years, this would mean a loss of lakhs of crores to national income. In contrast, even if the scheme was wildly successful in attracting inflows, the payout would only be in the tens of thousands of crores. Of course, there might be cheaper ways of restoring confidence but it was not clear what they were.

Discussions with my soon-to-be colleagues at the RBI threw up another rationale. The bankers who proposed the scheme were saying that if the money came in, the rupee would appreciate, and it would cost us less to offer the forward subsidy. This was self-serving and not quite right. What was right was that if we changed the narrative on India, and the rupee continued appreciating between the point when the money came in and the point we covered our liabilities to the bankers in forward markets, it would cost us significantly less. In fact, we might even make money on the deal. But what if the rupee plunged after the dollars flowed in? There were no certainties here.

The bottom line was that the scheme was a measured risk, with a probability that the RBI would lose money, a certainty that the bankers would make money, but also a reasonable chance that the country would be significantly better off. Having obtained the concurrence of the Finance Ministry, the RBI Governor had to decide.

Policy making is about deciding in the face of uncertainty, after weighing the alternatives as best as one can. On the day I was to take over, with no good options on the table, I had to choose the least bad one. I decided to go ahead.

The afternoon of September 4th 2013, we announced the package of measures. The FCNR(B) scheme drew in $26 billion, more than any of us anticipated. But more important, confidence picked up, the rupee continued strengthening beyond when the money came in, partly because the global investor mood as well as Indian electoral projections also changed, and we covered our forward swaps cheaply. We are fully covered for outflows today and have made money on the deal. The rupee has been one of the most stable emerging market currencies since then.

Of course, in hindsight it seems like it was the obvious thing to do because it worked. The reality is we will never know for certain! Perhaps it was the other elements of the package, perhaps it was everything together. Autobiographies are always written as if the author had it all mapped out with perfect foresight, ignoring the risks and uncertainties at that time. This misleads, as much as those beautiful photographs of a past holiday abstract from the heat, the mosquitoes, and the lack of connectivity. Policy making invariably involves taking measured risks in the face of uncertainty.

Clearly, we do not want to be in the position we were in August 2013 ever again. Macroeconomic stability is of paramount importance for India. Equally clearly, drawing from this experience, the central bank must have the resources, the knowledge, and the professionalism to act when the situation warrants. In the rest of this talk, I want to explain why this means India needs a strong and independent RBI to ensure macroeconomic stability. Then I will discuss what I think is needed to ensure such independence going forward.

The need for macroeconomic stability

Growth is good, but growth with stability is better, especially in a poor country where so many people live at the margin. For the RBI, it means ensuring growth does not exceed our potential, adopting prudential policies that reduce our risk, and building sufficient buffers that the country is protected against shocks.

Having your cake and eating it too: interest rates and the exchange rate

This mission, however, exposes the central bank to criticism. If we try and bring down inflation, interest rates will remain higher than borrowers desire. If inflation comes down, the currency will depreciate less than some exporters desire. If we push the banks to clean up, banks may be less tolerant towards habitual non-payers. Whatever we do, someone will object. The RBI then becomes the favorite scapegoat for underperformance - if exports are not picking up, it is because interest rates are too high and because the exchange rate is too strong.

Unlike the complainants, the RBI does not have the luxury of economic inconsistency. If we start buying dollars in a big way to depreciate the exchange rate, we will be able to buy fewer government bonds if we are to maintain control over liquidity. The consequence will be higher interest rates in bond market. Moreover, the depreciated exchange rate will mean higher inflation, which in turn will mean higher policy interest rates given the inflation objective the government has set for us. Once again, this means higher interest rates. Just look at Brazil or Russia to understand you cannot have a significantly depreciated exchange rate and lower interest rates at the same time if you want stable growth!

First year economics: there is no free lunch. RBI dividend policy

A fundamental lesson in economics is there is no free lunch. This can be seen in the matter of the RBI dividend: Some commentators seem to suggest that public sector banks could be recapitalized entirely if only the RBI paid a larger dividend to the Government. Let me explain why matters are not so simple. If what follows is complicated, trust me, it is. But pay attention, students, especially because it is about your money. I am sure you will understand.

How does the RBI generate surplus profits? We, of course, print the currency held by the public, as well as issue deposits (i.e. reserves) to commercial banks. Those are our fixed liabilities. As we issue these liabilities, we buy financial assets from the market. We do not pay interest on our liabilities. However the financial assets we hold, typically domestic and foreign government bonds, do pay interest. So we generate a large net interest income simply because we pay nothing on virtually all our liabilities.

Our total costs, largely for currency printing and banker commissions, amount to only about 1/7th of our total net interest income. So we earn a large surplus profit because of the RBI's role as the manager of the country's currency. This belongs entirely to the country's citizens. Therefore, after setting aside what is needed to be retained as equity capital to maintain the creditworthiness of the RBI, the RBI Board pays out the remaining surplus to the RBI's owner, the Government.

The RBI Board has decided it wants the RBI to have an international AAA rating so that RBI can undertake international transactions easily, even when the Government is in perceived difficulty - in the midst of the Taper Tantrum, no bank questioned our ability to deliver on the FCNR(B) swaps, even though the liability could have been tens of thousands of crores. Based on sophisticated risk analysis by the RBI's staff, the Board has decided in the last three years that the RBI's equity position, currently around 10 lakh crores, is enough for the purpose. It therefore has paid out the entire surplus generated to the Government, amounting to about Rs 66,000 crores each in the last two years, without holding anything back. This is of the order of magnitude of the dividends paid by the entire public sector to the Government. In my three years at the RBI, we have paid almost as much dividend to the government as in the entire previous decade. Yet some suggest we should pay more, a special dividend over and above the surplus we generate.

Even if it were legally possible to pay unrealized surplus (it is not), and even if the Board were convinced a higher dividend would not compromise the creditworthiness of the RBI, there is a more fundamental economic reason why a special dividend would not help the Government with its budgetary constraints.

Here's why: Much of the surplus we make comes from the interest we get on government assets or from the capital gains we make off other market participants. When we pay this to the government as dividends, we are putting back into the system the money we made from it - there is no additional money printing or reserve creation involved.1 But when we pay a special dividend to the government, we have to create additional permanent reserves, or more colloquially, print money. Every year, we have in mind a growth rate of permanent reserves consistent with the economy's cash needs and our inflation goals. Given that budgeted growth rate, to accommodate the special dividend we will have to withdraw an equivalent amount of money from the public by selling government bonds in our portfolio (or alternatively, doing fewer open market purchases than we budgeted).

Of course, the Government can use the special dividend to spend, reducing its public borrowing by that amount. But the RBI will have to sell bonds of exactly that amount to the public in order to stick to its target for money creation. The overall net sale of Government bonds by the Government and the RBI combined to the public (that is, the effective public sector borrowing requirement) will not change. But the entire objective of financing Government spending with a special RBI dividend is to reduce overall Government bond sales to the public. That objective is not achieved!

The bottom line is that the RBI should transfer to the government the entire surplus, retaining just enough buffers that are consistent with good central bank risk management practice. Indeed, this year the Board paid out an extra 8,000 crores than was promised to the Government around budget time. Separately, the government can infuse capital into the banks. The two decisions need not be linked. There are no creative ways of extracting more money from the RBI - there is no free lunch! Instead, the Government should acknowledge its substantial equity position in the RBI and subtract it from its outstanding debt when it announces its net debt position. That would satisfy all concerned without monetary damage.

If what I have said just now seems complicated, it is, but it is also the correct economic reasoning. Similar detailed rationales lead us to turn down demands to cut interest rates in the face of high inflation, to depreciate or appreciate the exchange rate depending on the whim of the moment, to use foreign exchange reserves to fund projects, to display forbearance in classifying bad loans or waived farmer loans as NPAs, and so on...

We have been tasked with a job of maintaining macroeconomic stability, and often that task requires us to refuse seemingly obvious and attractive proposals. The reason why we have to do what we have to do may not be easy for every unspecialized person, even ones with substantial economics training, to grasp quickly. Of course, we still must explain to the best of our ability but we also need to create a structure where the public trusts the central bank to do the right thing. This then is why we need a trusted independent central bank.

Central bank independence

In this environment, where the central bank has to occasionally stand firm against the highest echelons of central and state government, recall the words of my predecessor, Dr. Subbarao, when he said "I do hope the Finance Minister will one day say, 'I am often frustrated by the Reserve Bank, so frustrated that I want to go for a walk, even if I have to walk alone. But thank God, the Reserve Bank exists.'" I would go a little further. The Reserve Bank cannot just exist, its ability to say "No!" has to be protected. At the same time, the central bank cannot become free of all constraints, it has to work under a framework set by the Government. This requires a number of actions.

Outline responsibilities of the RBI

When the responsibilities of the RBI are fuzzy, its actions can continuously be questioned. Instead, if the constitutional authorities outline a framework for the responsibilities of the RBI, it can take actions consistent with those responsibilities and be held to outcomes. The inflation objectives recently set for the RBI by the Government are an example of what is needed. Critics can lambast the RBI if it fails continuously to meet the objectives, but if they want it to lower interest rates even when the RBI barely meets its objectives, they should instead petition the Government to change the objectives.

Similarly, the RBI Board has adopted a risk management framework which indicates the level of equity the RBI needs, given the risks it faces. The dividend policy of the RBI then becomes a technical matter of how much residual surplus is available each year after bolstering equity. Frameworks thus reduce the space for differences.

The RBI's role in macroeconomic stability is, however, still fuzzy. While RBI clearly has responsibility for the safety and soundness of credit institutions and the stability of the external account, there are some areas that are hazier. For example, with an inflation focused framework, the RBI's ability to be accommodative depends on fiscal prudence from the center and states. How much should the RBI warn on fiscal profligacy, including the building up of contingent liabilities, and when should such warning be seen as interfering in the legitimate decisions of the elected representatives of the people? This is an area where clarity would be useful.

Strengthen oversight

The freedom the RBI has to take operational decisions such as the FCNR (B) swap arrangement, albeit invariably in consultation with the Finance Ministry, is important. However, there are always government entities that are seeking oversight over various aspects of the RBI's activities. Multiple layers of scrutiny, especially by entities that do not have the technical understanding, will only hamper decision making. Instead, the government-appointed RBI Board, which includes ex-officio government officials as well as government appointees, should continue to play its key oversight role. In this regard, all important RBI decisions including budgets, licenses, regulation, and supervision are now either approved by the Board or one of its sub-committees. Vacancies in the RBI Board, which have remained unfilled for many months now, should be filled quickly so that the full expertise and oversight of the Board can be utilized.

It is also important that Parliament understand what the central bank is doing. The Governor and Deputy Governors interact regularly with various parliamentary committees, but we have also initiated a six-monthly interaction with the Parliamentary Standing Committee on Finance, where the Governor reports on the activities of the Bank, and the Committee offers its views and concerns.

Rank of RBI Governor

There is a reason why Central Bank Governors sit at the table along with the Finance Ministers in G-20 meetings. It is that the Central Bank Governor, unlike other regulators or government secretaries, has command over significant policy levers and has to occasionally disagree with the most powerful people in the country.

It is dangerous to have a de facto powerful position with low de jure status. Today, the RBI Governor has the salary of the Cabinet Secretary. He or she is appointed by the Prime Minister in consultation with the Finance Minister. The Governor's rank in the government hierarchy is not defined but it is generally agreed that decisions will only be explained only to the Prime Minister and the Finance Minister. There is an informal understanding in India that the Governor has the room to make needed decisions. In the interests of macroeconomic stability, none of this should be changed, though if these issues are ever revisited, there may be some virtue in explicitly setting the Governor's rank commensurate with her position as the most important technocrat in charge of economic policy in the country.


I do not fool myself into thinking that reporters and TV cameras follow me around everywhere because I am a magnetic speaker. They follow RBI Governors around because they may offer market-moving information on policy. Fortunately, I have escaped unintentionally saying anything that moves markets.

At the same time, while different RBI Governors have different approaches to communication, no one can dispense with it. Unlike a developed country, where the Central Bank Chairman can offer periodic Delphic pronouncements about the course of monetary policy, and occasionally remonstrate with Parliament or the government about the course of fiscal policy, in our developing democracy the RBI Governor has to continuously make the case for the actions the central bank is taking, including the many structural reforms that are underway.

Indeed, communication is as much about educating as it is about informing. For instance, even as I explained to entrepreneurs and retail borrowers why interest rates were not falling faster, I had to use the price of dosas to explain to pensioners why they were actually better off earning lower nominal but higher real interest rates. Public understanding can help ease the way for reforms, as well as increase support for policies. The RBI Governor therefore has to explain again and again.

Occasionally, of course, the Governor has to warn about the dangers of certain courses of action or certain tendencies in the economy for growth and macroeconomic stability. Finally, the Governor is also a role model for the youth in this country, and should therefore not duck the responsibility to urge them to follow the highest standards of citizenship when he or she is invited to speak directly to them.


My parents could give me a wonderful childhood and a great education in this country. We all grow up with such debts, and public service is a way to partially repay them. However, the last few years, working with dedicated colleagues at the RBI and the Finance Ministry, have been so fulfilling that I think I have received rather than given. I believe we have undertaken important reforms in payments, in banking, in the conduct of monetary policy and liquidity management, in financial markets, and in the resolution of distress, as well as within the RBI itself. Only time will tell whether they will have lasting impact, but I tried to do the best job I could, without fear or favor. This is the last public speech I will give in India for a while - my successor has to take over the RBI's communication and I want to get out of his way. It has been an honor to work for the country and especially to talk to people like you, its future. Thanks for listening to me.

1  This is not strictly true. Our earnings on foreign exchange assets come from outside the system, so when we pay this to the Government as dividend, we are printing additional money. We do account for this.