Search interesting materials

Wednesday, October 02, 2013

Who's afraid of a big current account deficit?

A big CAD is a bad thing -- much like a big fiscal deficit.

A country is always better off with a small or zero CAD or ideally a surplus.

The CAD is a drag on growth.

The large CAD is a profound drag on India's outlook.

If we managed to reduce the CAD, things would get better.

Statements like this are rife. They are wrong.

What is the CAD?

The CAD is three things, all of which are identical. It is the gap between revenues from selling goods and services versus the payments made for buying goods and services. This has to be exactly matched by the capital inflow into the country. This is exactly equal to the gap between investment and savings. These three relationships are accounting identities.

What if there was no capital account?

If there was no capital account, then the proceeds from selling goods and services would have to exactly match the payments for goods and services, in every minute. Every small mismatch between the two would generate extreme currency fluctuations (large enough to incite a current account response).

The capital account is what smooths these things out. Let us imagine the currency market for one minute in which someone is buying \$1 billion in order to import something. In that very same minute, it is very unlikely that there will be a double coincidence of wants, in the form of an exporter who wishes to sell \$1 billion. What fills the breach is the capital account. Some speculator comes in and supplies that \$1 billion in the hope of scoring a short-term speculative profit. The real economy demands liquidity in the currency market and finance supplies this, through the capital account.
The CAD is exactly equal to the gap between savings and investment. A CAD of zero is tantamount to only investing what we have saved. In general, this is a bad idea. If the country is all set to invest 35% of GDP, and savings are only 30% of GDP, it is a good thing if capital flows of 5% of GDP show up, through which investment exceeds savings.

Should we bemoan the large Indian CAD?

Should you be unhappy that investment is bigger than domestic savings by 5 per cent of GDP? If we insisted that the CAD should be 0 (i.e. we had no capital account) then investment would have to be lower and savings would be higher (which in turn implies reduced consumption). This would give reduced GDP growth.

Financial autarky implies that we live within our means. With savings of 30% of GDP, investment is forced to 30% of GDP under autarky. Opening up to the world makes it possible for a country to import or export capital.

If a country has good prospects but low savings, running a CAD is a way to front-load the investment, and service the foreign capital through a stream of dividends, interest payments and debt repayments into the future. If a country has poor prospects, it is better off sending capital to good uses overseas, instead of investing it domestically. For these gains, we have to have an open capital account and run large and variable CADs.

(There are also gains from risk sharing from large gross capital flows, even if the CAD is 0, but that's a separate topic of discussion).

A big CAD got us into trouble in 1991. Won't that happen again?

In 1991, FERA (1973) was in force. Capital account transactions by private parties had been criminalised. The only mechanism that generated flows on the capital account was the government. The entire CAD had to be financed by government borrowing. When the government lost creditworthiness in the eyes of the world, we had a funding crisis on the capital account.

On a day to day basis, imports required dollars which came from the government. The Ministry of Finance monitored daily inflows and outflows of dollars, and controlled who could access foreign exchange.

When GOI lost credit-worthiness in the eyes of overseas lenders, this was a collapse in the flow of dollars. If you wanted to import penicillin, you needed to get dollars, and RBI had none. That's where it came to crunch: when an importer is told that he cannot import as there are no dollars.

Nothing remotely like this can happen in the present environment. With capital account liberalisation, many channels have opened. There is FII investment in equity and debt, there is FDI, there is ECB, and so on. The money moving in these channels dwarfs the borrowing by the government. India is now well connected into financial globalisation. All these channels won't choke.

Suppose there is some big mess abroad and all fixed income funds stop buying Indian bonds. Under these circumstances, capital inflow will come through the other channels. The more we open up to a diverse array of investors into a diverse array of asset classes, the safer the environment becomes, the lower the exchange rate volatility becomes.

Why won't all channels choke all at once?

We require a capital inflow, on average, of Rs.20 billion per day. That's the gap, on the currency market, which has to be filled. If foreign capital does not come in, there is a supply-demand mismatch on the currency market. This gives a currency depreciation.

Ex-post, supply always equals demand. On the market, this demand will be met. Every day, the CAD of the day will equal the capital inflow of the day. The only question is: At what price?

When bad news comes out in India, foreign capital becomes more circumspect. They require a more attractive exchange rate at which to get in. Or, to say it differently, suppose INR/USD is at Rs.65 to the dollar. Suppose bad news come out. The inflow of Rs.20 billion is not forthcoming. The market has a gap. The rupee starts falling. At Rs.70 to the dollar, some foreign investors think `Hmm, maybe at this price, it's a good deal, and I should get in'.

How far does the depreciation go? Minute by minute, the rupee moves to elicit the net capital inflow (or outflow) required to clear the currency market. In response to bad news, the INR drops till a speculator feels that it might be a good idea to come into India, buy a 91 day treasury bill, and hope that the rupee will do well in a few minutes or few days. That's how the current account deficit always gets financed under a floating exchange rate.

Rupee depreciation makes Indian assets more attractive. It would be nice if foreign capital found Indian assets attractive for other reasons. But when all else fails, rupee depreciation is what gets the job done.

What kinds of foreign investors respond the most to rupee depreciation?

Sharp spikes of the rupee are fertile ground for currency speculators. The more currency speculators that we have, who are operating on the rupee market, the smaller is the INR movement associated with an event.

Imagine an INR depreciation of 5% in one day. A currency speculator believes this is over done and wishes to come in. What does he do? He sells dollars, buys INR, and invests in short-dated government bonds. This would add up to a pure play on INR. Currency speculators are not comfortable holding Nifty in India. They want a pure exposure to INR.

Hence, the best way to obtain a deep and liquid currency market, where shocks will lead to small exchange rate fluctuations, is to remove capital controls on the rupee denominated debt market.

A big CAD increases the damage caused by a sudden stop in capital inflows. What should the country do to forestall this?

Sudden stops are ultimately about asymmetric information in the hands of foreign investors. If India has a deep engagement with financial globalisation, then the informational asymmetry will be removed.

Our policy goal should be to have thousands of global financial firms who are running business activities connected with India, who have large scale organisational and human capital that is devoted to understanding India. This deep engagement will deter problems such as home bias, sudden stops, etc.

The Indian capital controls are damaging this deep engagement. As an example, repeated stop-go policies frustrate the development of teams inside global financial firms that have deep knowledge about India. When these teams know less about India, there is a greater likelihood of encountering the pathologies of international finance.

When India does silly things like trying to `crush the speculators' through various means fair and foul, this hinders a mature engagement with financial globalisation. When global capital feels that India operates on stable rules of the game and has mature policy makers, the resources committed for building organisational capital connected with India will be greater.

In order to avoid international finance pathologies such as sudden stops, our engagement with financial globalisation should be a deep engagement. While this issue becomes particularly salient when the CAD is large, but there is no short term solution. Over the years, we have to chip away at building a deep engagement with financial globalisation at all times, so as to reduce the risk when there is a large CAD.

This is like a rules versus discretion problem. When discretion is used at a time of a large CAD, it contaminates credibility at all times. A mature approach to public policy involves establishing capable institutions that implement stable rules of the game and not tactical dogfights.

What is the role of MOF or RBI in ensuring adequate capital comes into the country to match the CAD?

On a day to day basis -- nothing. It's a purely market process. The market does it. There are no gray men who look at the CAD and figure out how to finance it and then undertake actions through which it gets financed. The financing of the CAD is purely a market process.

The role for MOF and RBI is to get out of the way by removing capital controls, so as to reduce the magnitude of INR depreciation required when a certain negative event takes place.

Does this work differently for other countries?

A large CAD is dangerous when there is a managed exchange rate. Under a managed exchange rate, there is a propensity to borrow in foreign currency and leave it unhedged. These borrowers (whether corporations or governments) get into big trouble when there is a large exchange rate depreciation.
The central bank is much more likely to fail on exchange rate management when there is a large CAD.

The witches' brew that adds up to trouble is a central bank that believes there should be exchange rate policy + borrowers who believe the central bank will pursue exchange rate policy + a large CAD.
While India has a de facto floating exchange rate, RBI has not yet stopped talking about dreams of exchange rate management. We are relatively safe because borrowers don't believe RBI can do much about the exchange rate. Hence, there is no moral hazard and a large CAD poses no threat.

Why is a large CAD seen as a big problem?

With a large CAD, India is beholden to foreign capital inflows. If foreign investors are displeased, we get a big rupee depreciation. This generates accountability.

When India enacts capital controls, or the Food Security Bill, we get a rupee depreciation. This irritates policy makers, who feel that mirrors should reflect a little before throwing back images.
Nobody likes accountability. Hence, people in positions of power do not like a large CAD.

In a mature market economy, a key channel of accountability for the government is the bond market. When the government does bad things, their cost of financing goes up, and this directly hits the ability of politicians to spend on their pet projects. In India, the bond market has been muzzled by setting up a system of financial repression. The job of intimidating the authorities is then left to Nifty and the rupee. The voice of the latter is amplified when there is a large CAD.

If you look at the world from the viewpoint of the people who run the place, there is a desire to muzzle Nifty and the rupee (particularly when the latter is speaking loudly thanks to a large CAD). From that viewpoint, a large CAD is a bad thing. Because the establishment has a disproportionate impact upon the climate of ideas, we have started accepting their claim, that a large CAD is a bad thing.

If you care about India's future, a large CAD is a good thing, as it enhances accountability. By this logic, other things being equal, the Indian policy process generates superior outcomes when there is a large CAD. If we had a small CAD, Mr. Mukherjee might have been finance minister today.


Financial globalisation is work in progress. Capital controls and source-based taxation hinder international capital mobility. Even if there are no restrictions, it is hard for investors in country i to properly utilise the investment opportunities in country j, for reasons of `information distance'. All too often, there is home bias (people in a country holding vastly greater domestic assets than is optimal from the viewpoint of diversification). There are international finance pathologies such as capital surges, sudden stops, investments by foreigners in wrong assets, and so on. These are the hurdles along the road.

In the destination state, there is no good reason why the investment opportunities in country i at time t should match the savings of country i at time t. We should judge the success of the project of financial integration by the extent to which we are able to achieve large and variable current accounts.

In addition, in a place like India, a big CAD generates greater accountability on the part of the government. One would predict better economic policy when there is a large CAD.

The widespread mistrust of a large CAD may reflect two things. Some don't see the extent to which we're not in 1991 anymore: there is much more of a deep engagement with financial globalisation, and the exchange rate floats enough that the borrowers are not unhedged. And, establishment figures resent accountability.

I am grateful to Josh Felman for illuminating discussions on these issues.


  1. When rupee depreciates, indian assets earning rupees won't change profitability.
    If asset price goes down in dollar terms, earnings also go down in dollar terms.
    So how does rupee depreciation make assets cheaper for outsiders?
    Are we confusing currency calls with asset valuations?
    Or I have a vacuum somewhere in the way I have understood it?

    1. A sharp rupee depreciation offers a possibility that INR has overshot. If so, a speculator who comes in to buy INR will get a good return in a short time when the rupee comes back closer to fair value.

      Walking is controlled falling.

  2. Quite remarkable. Please try and get this published somewhere, such as American Economic Review. This is the stuff that makes our academics in our great public institutions such great economists, universally lauded for their insights worldwide. I enjoyed it immensely. Such wisdom is rare. Thanks.

  3. You are distorting one of the basic economic identities to suit your arguments. I challenge if this can go beyond mere a "blog" entry and stand up to scrutiny of any decent journal. But then the purpose here is to influence/fool the general public.
    If a country is running current account deficit then that's needs to financed by capital account surplus which is nothing but "selling" of domestic assets either held abroad or inside the country's boarders.
    If the capital was freely flowing, as you make it to be, without any "ownership" or any "nationality" attached to it then we would not be discussing any current and capital account in the first place.
    The trick employed by apologists of foreign ownership of domestic assets like you (in the form of "CAD is such a good thing") is you completely camouflage the ownership angle as if whether the Indian assets are owned by Indians or foreigners doesn't matter at all.
    But the cake will take the "accountability" argument. It's like arguing that if I own a house I will not be accountable for it's maintenance but if I take a house on rent I will always be accountable to the owner of the house and that's such a good thing!!

    1. FDI isn't just about 'owning' domestic assets, but investment into creating new domestic assets (which might not even be physical assets, but IP, etc) which make a return for the foreign owner and the local economy. So, it depends on other rules and policies to do with FDI. That being said, I would prioritize domestic investment over FDI. Today, domestic investors are also not investing due to various problems and are going abroad, so those problems should be prioritized.

      We often tend towards xenophobia because of our colonial past. But, if one looks at many MNCs, they tend to have better governance and technology and quality standards than local business and we get to import technology and know-how which is pretty standard in the rest of the world. So, FDI has a lot of benefits as long as your concerns are reflected by adequate protections on the investment made into domestic assets. Furthermore, foreign capital can demand greater accountability from the government as opposed to domestic capital (which is boxed in and has no choice). When the government is the bottleneck as it is today, that is a good thing.

    2. Thank God !! Raghu is as xenophobic as I am.

  4. Economic well being is not equal to overall economic growth. There is the question of inflation and how individuals are impacted by a set of policy options. Depreciation of Ruppee (with fall in capital flows) increases inflation and disadvantages certain people e.g., who consume a lot of petroleum products, kids wanting to go overseas for education, people whose incomes are fixed, etc.

    Q for the author -- 5 people are earning Rs 20 each (Rs 100). As a result of a policy, 4 people start earning Rs 15 each and one starts earning 50 (total of 110). Overall impact is an increase in income of 110. However, 4 are negatively impacted and 1 benefits. Should the policy be implemented or not ?

    Another question for the author -- if CAD was to increase to say 100% of the economy, no capital would flow. Question for you what is a comfortable level of CAD ?

    Actually the reverse happens in Petro states -- Countries export oil and use the earnings to fund expenditure. When prices for oil fall -- people's income come and down and puts a huge pressure on the economy. This happenned to esrtwhile Soviet Union. We are opposite, relying on capital flows -- whenever it falls (we are squeezed).

    1. Two things:

      The rupee depreciation is not a root cause as you make it out to be. For rupee to appreciate we need to fix the root causes of low productivity and low competitiveness. It makes no sense to ask for a stronger rupee without asking to fix the root causes which need long term commitment and a major change in the way we function.

      Secondly, its not clear how many people benefit and how many don't. Rupee depreciation helps exporters and pretty much every company benefits by being more competitive in comparison to other countries. IT industry benefits a great deal. All of that leads to more jobs in the export services and manufacturing sector. Although, its not as much as one would hope for because of other hurdles put up by poor government policies. But, because of the examples I gave, I am not sure whether we have a net benefit or loss.

      Furthermore, it is a self-corrective system. A lower currency would promote more investment into exports which eventually would increase productivity and lead to currency appreciation. That is, if we implement the right policies to increase productivity which is the only way to have a strong economy and currency, at the end of the day.

    2. Agree that the economy needs to be more efficient and competitive and that the value of Ruppee is an outcome of the relative strenth of the Indian economy and not a cause.

      There are arguments for both a higher and lower Ruppee. As you point out lower Ruppee makes products and services competitive but reduces the buying power of the citizens. There are tradeoffs involved. Point is not high or low Ruppee but stability in the value and not the high level of volatility as has been witnessed. Volatility makes planning very difficult.

      Capital flows can slow down for a number of reasons --both internal and external. High level of CAD makes the Ruppee vulernable to high level of volatility. What happens if the capital flows remain low for an extended period of time, Ruppee goes to 100 and petrol sells for 115 Ruppees ?

      Self correcting mechanism is good in theory, but can be a challenge in situations of high volatility e.g., during the GFC, should the banks have been allowed to fail and go out of business ?

      Point being made was that the article argues the issue from one point of view and overlooks the trade offs involved in having a high CAD.

    3. Well, much of the problem accumulates over times of low volatility, when the problems are ignored for long periods. The actual periods of volatility are actually simply resolution of long-standing issues. The period of volatility was great from the point of view of long term fundamentals as billion of dollars worth of projects got cleared, many measures were taken which were long pending - like clearance for foreign universities and FDI for other sectors. If we get the kind of urgency we saw from the govt in pushing through policies during this volatility, I'm prepared to have rupee go to a 100 and petrol at 115. No problem, whatsoever. :) Just imagine how much will get done by the govt if that happens!!

    4. Proof of the pudding lies in eating it. So a suggestion -- you could argue to the Government to have policies that will lead to a higer level of CAD, as this will drive down Ruppee, spur growth and make Indian goods and services more competitive. Make the suggestion and find out whether you premise is implementable or academic ?

    5. One more comment -- policy announcements and project clearances are not equal to inflows and investment in projects e.g., no FDI in retailing till date.

    6. Why would I argue THAT to the government? And, why put the cart before the horse? Even if I had to argue anything to the government I would argue in favor of long term policies. The CAD is irrelevant, its a secondary measure.

      Yep, very little FDI in retail. Which means that the policies need further work. Precisely why one gets the sense that more pressure is needed from the market. By the way, I have no desire for FDI. It would be nicer if it were domestic investment but it seems domestic business is not willing to play ball with the govt given its arbitrariness. And, similarly for FDI. No wonder there is no FDI in retail. That's exactly the point! Another example is the policy for foreign universities - in 2008 the goal was a grand opening up, and it has been opened up in small steps, liberalizing marginally every time because whatever has been done hasn't been enough. 5 years and counting... eventually, we will get to the right policy for investment into higher education and similarly for retail - domestic or FDI or whatever....

    7. Thanks for taking time to respond. Do not agree with your thesis. Will sign off with 2 thoughts:

      1. Should you include some real quantitative data to support the thesis ? Otherwise, one could argue that the article is based on the author's subjective logic.

      2. Should the tradeoffs be discussed i.e., impact of CAD on a number of economic indicators on both the short term and the long term ?. analogy -- while prescribing a medicine, should you also consider side effects/ other intended effects.

  5. The statement " CAD is exactly equal to investments minus savings" is quite misleading and partially incorrect. I believe budget deficit is the primary mechanism to support the gap between investment and savings. CAD is only a secondary source of support , only to the extent of the need for importing capital assets (machineries, intellectual properties etc).

    I think you should have a follow up blog to clarify your thought process.

    1. Ajay, you crack me up.

      Anyway, I think the comment above it on internal vs external drivers of capital flows is the more interesting one.

      In general your piece does a good service, and I'll likely give it to students to read. But it appears premised on the standard idea that capital responds mostly to domestic conditions. There's a lot more room for legitimate argument when K flows are externally driven (e.g. Helene Rey's recent piece

      In India's case, this isn't to say the rupee didn't deserve to get hammered. Rather, the external forces of the QE-driven K inflows probably prevented it from getting a much-deserved hammering earlier. So India suffered because of not getting the accountability signal earlier and having to tolerate Mr. Mukherjee longer than it should have.

  6. This sort of lucid and reason based thinking is almost heresy in Indian economic policy establishment. I would like to echo the sentiment expressed in a comment earlier ... This sort of analysis and wisdom is rare.
    The point on the Bond market being the barometer of government performance in advanced economies is spot on.

    1. This was an excellent article, but I dont agree with the central theme of the argument that a large CAD is good. To understand your thought process clear, I would be extremely grateful if you can answer my following queries:
      1) Consider situation where CAD is 0 or even surplus...Still foreigners eg. PE funds, FIIs investing in Indian bond and stock markets with the confidence in the economy...Wouldn't investors be more comfortable hoping rupee would appreciate and they gain thereby and not just making an opportunistic bet when rupee depreciates
      2) Doesn't your argument make implicit assumptions of 1) Elasticity of exports and imports 2) Perfect mobility of capital flows 3) Rupee remaining at the same or stable level throughout the process 4)Prepondarance of only short term investors, rather speculators....None of them hold true in real life
      3)In the Savings, Investment, CAD identity, are we not ignoring the role of Fiscal deficit....A high level of investment may co-exist with a high fiscal deficit leading to a problem of twin deficit. Is it correct to portray an identity as it were a functional relationship. It can also be argued that a 0 CAD could co-exist with Investment exceeding Savings and balance getting adjusted in reduction of Fiscal deficit (Utopian, I know, but just a possibility which negates the statement that CAD is good.

      4) I feel that your argument about accountability is spot on, so is the point about financial market repression...but dont you think that import of gold which worsens (or in your view betters) the CAD situation, is definitely bad for the economy

      5) If I understand correctly, Investment in the Investment saving identity means real investment or investment in physical does investment in financial assets (which is not a part of GDP and hence the identity you have referred) affect the Real investment Savings gap? Isn't investment in bonds by foreigners just a 'short term financing' of the rupee....some kind of succor given to the rupee which the rupee will get addicted to, in CAD persists, causing the currency forever to be dependent on the foreign flows...Ideally should it not be the other way round?
      What has this to do with real investment anyways?

      Thanks very much for the article...It was extremely though provoking and stimulating


      Kunal Khairnar


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticism is perfectly okay; uncivilised language is not. We delete any comment which is spam, has personal attacks against anyone, or uses foul language. We delete any comment which does not contribute to the intellectual discussion about the blog article in question.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.