The Impossible Trinity

Economists who are specialized in macroeconomic analysis of open economies like very much to mention several times the so-called “Impossible Trinity”, which was demonstrated in the sixties and seventies – still during the times of fixed exchange rates – by Nobel Prize Robert Mundell as well as other economists such as Fleming, Corden, Johnson and Fukuyama.

Basically, it was clearly proved by them that it is not possible at the same time to have full and free international capital mobility, a fixed exchange rate, and an independent monetary policy. In other words: one of the three things has to be modified: either one reduces capital mobility, or one adopts a flexible exchange rate, or monetary control must be abandoned.

Let us consider, for example, the text below published recently in a “Morgan Stanley Economic Bulletin” about India:

“Fighting the Impossible Trinity”

“The Reserve Bank of India is increasingly facing the complex challenge of trying to pursue an open capital account, an independent monetary policy, and a managed exchange rate. Given that India has gradually opened up its capital account, the Central Bank is being forced to choose between the other two corners of the impossible trinity – i.e., an independent monetary policy and a stable exchange rate.”

However, what happens is that from the early seventies on, but particularly now in this new century, when capital mobility really increased intensely, we are somewhat discovering a “new” impossible trinity: under high international capital mobility, independently of the foreign exchange regime which is adopted, it is not possible to have anymore an independent monetary policy.

In other words: independently of the exchange rate regime, one of two things must be modified: either one restricts capital mobility or one loses control of monetary policy.

What we mean, with this “new” vision of the impossible trinities, is that – even under floating exchange rates – whenever there is great international capital mobility, the control of monetary policy is lost. The nominal interest rate which would be the best for a policy of antinflationary targets becomes, as a matter of fact, the worst nominal interest rate for economic growth, as well as for the real effective exchange rate. The action of the Central Bank on interest rates has great influence on the inflow or outflow of international capitals, either with fixed exchange rates or with floating exchange rates.

As a matter of fact, from an operational point-of-view, the traders of the new century simply ignore the possibility of exchange rate devaluation, and this causes a significant interest rate differential to have the same impact under fixed exchange rates or under floating exchange rates. Even worse: a favorable interest differential tends to provoke an enormous appreciation of the nominal exchange rate, according to a phenomenon called “overshooting” by Rudiger Dornbusch during the seventies – an appreciation which goes up to a certain determined extreme point so ultra-overvalued in which the market might really begin to consider – only departing from that point – some devaluation expectations for the clearly overvalued exchange rate.

The famous theory of an independent monetary policy – supposedly even allowing the establishment of certain inflation targets – falls apart, to the extent that the very policy decision of a major interest rate change begins to exert an immediate effect over the exchange rate itself, producing this overshooting impact.

The Central Bank is going to influence inflation via interest rates through the slow transmission channels of monetary policy, but it affects the exchange rate – and much more immediately.

In the Brazilian case, inflation sometimes tends to stay even below the targets, and one might even see some deflation in reais in many products and services, but the exchange rate becomes extremely overvalued. Therefore, monetary policy is not independent of the exchange rate and cannot be focused on the rate of inflation only.

At the end of the day, the basic dilemma (or trilemma) remains: either you move the interest rate because of inflation preoccupations or you act over the interest rate due to the exchange rate consequences.

And all that without mentioning the serious problem of unemployment and recession, which in fact can never be ignored by Central Banks, even when they state firmly that a monetary policy regime should only focus on inflationary targets.

Summing up: in Brazil, either one reduces immediately interest rates to 8 or 7%, or one establishes taxation or controls over short term international capital movements. James Tobin – also a Nobel Prize economist – suggested the so-called “Tobin tax” and used to say that: it is necessary to put some sand on the windmills of international financial globalization. Without lowering interest rates and without taxes or controls on the capital account of the balance-of-payments, the dollar will tend to depreciate even more against the Brazilian currency (real), in the direction of 1.5 or less. One of the worst “Dutch diseases” in Brazilian History. Worse than 1995-1998.

7 Responses to "The Impossible Trinity"

  1. Nelson Noya   May 16, 2008 at 10:08 am

    As you said, the trilemma is based on the full international mobility of capital flows. But this condition implies that domestic currency denominated assets are perfect substitutes of FX assets. Do you believe so?Additionaly, perfect substitution implies UIP (uncovered interest rate parity) wich usually does not hold, at least in short horizonts. UIP is one of the most known puzzles that defies "textbook" simple macroeconomics.So, why are so many people still asuming trilemma, when there is no empirical foundations for it?

  2. Lemgruber   May 16, 2008 at 10:43 am

    Nelson NoyaThis is the puzzle. The traders are looking for high interest rates. The funny thing is that the interest rate parity generates in the futures market always an expected devaluation for the real, but it never happens. This only helps because the future price of the real is even better.Lemgruber

  3. Vitoria   May 16, 2008 at 12:02 pm

    Professor,While I agree with you, don’t you think that the actions from last year when the Central Bank was buying dollars it was the righrt policy?

  4. JB   May 17, 2008 at 2:51 am

    Dear Antonio,I very much appreciate your analysis, and also want to take it a step further.You state that Central banks have an issue with influcencing the monetary markets in the medium run as they utilize interest rates which is a long term adjustment tool and due to interconnection of markets is now becoming a tool with immediate effects to money markets exchange rates and economic growth.Now, while this is absolutely correct, this is like a doctor decribing a symptom but not the diseas. The underlying issue is that money policy is influenced mainly by two factors beyond the control of central banks:First the taxation rate, which as you state as well is under control of the government. As a result it is necessary that Governments act following a Central Banks interest rate decision. In our days however, we see the opposite. Governments e.g. (France) try to act actively and openly – even threatening – the Central Bank into a desired direction. That is absolutely counterproductive to the desired outcome of a stable currency and economy.Second, and this is the largest mistake economic theory has been doing – measuring comsumption as a factor in the GDP, as a factor of productivity. That has a good fundamental basis, but once an economy like the US is changing towards credit based consumption, the underlying fundamentals are gone.Instead of using Consumption we should use an NPV adjusted consumption in the GDP. And here you are right again, it causes the whole world a large headache, because nobody really knows the true value of the currency.Best regardsJB

  5. lemgruber   May 17, 2008 at 8:18 am

    vitoriamy major emphasis is to argue that there is nothing heterodox about a tobin tax on capital flows. i am trying to argue against the argument that a tax like that is not compatible with a modern monetary policy regime with floating rates and inflation targets.lemgruber

  6. ewulf   May 17, 2008 at 6:48 pm

    The core of the argument is the level of integration of capital markets and substitution between domestic and foreign assets, all of which are long run conditions. In the short run,Monetary policy effectiveness also depends upon the complemantary fiscal policy at works,such that it might allow more room to works on inflation without hurting too much the long term value of the exchange rate.

  7. Mohit Satyanand   May 18, 2008 at 8:25 am

    Ove the last 8 weeks, despite a stable-to-hawkish monetary regime, and high interest rates, the Indian currency has witnessed rapid depreciation, showing that the carrot of high interest rates is not enough to offset the risk of devaluation. In the context of a swelling Indian oil import bill, this is probably the rational response, showing that at least some of "the traders of the new century" have a more nuanced approach towards interest rates