More IMF public support for capital controls

Following up on maladjusted's weekend update, Nicolas Eyzaguirre, the IMF's top guy for the western hemisphere, said yesterday that the Fund supports the use of controls on capital inflows to prevent unwanted currency appreciation and potential overheating. Now, this isn't breaking news, but it's certainly nice to see someone high up reiterate this position, especially after the Global Financial Stability report released a couple weeks ago seemed to backtrack a bit.

In any case, Mr. Eyzaguirre had this to say:

In some cases, exchange rate flexibility, good fiscal discipline, and a prudent set of macro prudential policies, may not be enough to avoid an over-expansion or a potential bubble. Thus, we do not bar or disapprove the potential contribution of carefully designed taxes on capital inflows that may have a role in complementing the policy toolkit—although we should always bear in mind that these kinds of more orthodox responses do have their limitations.

And he should know too, having worked at Chile's Central Bank during the 1990s when the country implemented its own well known capital controls.

The IMF's new position on capital controls is by no means groundbreaking. It actually follows the consensus in the academic literature quite closely and can essentially be summed up with this nice flow chart (click to enlarge):

If you face a surge of capital inflows (as many developing countries are facing today due to low interest rates in the north) you should first consider if your currency has room to appreciate. If this isn't the case, then you should consider if reserve accumulation is desirable and if inflation is a problem, then you should attempt sterilization. But sterilization can prove too expensive, in which case capital controls become a useful policy option. Of course, if inflation isn't a problem you can just lower interest rates to discourage the inflows from coming at all.

This is all very right, but there are certainly a lot of caveats in the IMF's position--enough to make one wonder how often the Fund will actually recommend the use of capital controls during its consultations with its member countries. For instance, the Fund insists on overstating the problem of the evasion of controls by sophisticated financial actors and has also warned that controls cause distortions.

Now, of course any regulation is only as good as its regulators. But nevertheless, no matter the extent of sophistication of your financial sector, evading controls always entails a cost, which is exactly what controls are supposed to do. In fact, most empirical studies have found controls effective to the extent that they manage to "segment" the domestic financial market from the international market--meaning there's a wedge between domestic and international interest rates or between the price of stocks traded in both markets. And very frequently, this is the case even in countries where critics have claimed that controls were evaded.

IMF backpedals its endorsement of capital controls

Maladjusted's readers may know that after being snubbed by developing countries because of its horrible policy advise, the IMF had a midlife crisis and was forced to do a little soul searching. But right when it was on the brink of utter irrelevance, the global financial crisis hit and the ailing institution was given a new lease on life. In the process, headed by french "socialist" DSK, the IMF began reevaluating many of its long held dogmas.

Most recently, the IMF came out in favor of using short-term controls on capital inflows. This was considered a BIG FUCKING DEAL, and rightly so, but the details of how this would translate into actual policy were anything but clear. So now it shouldn't exactly come as a surprise that the IMF seems to be softening it's support for capital controls--that is, taxes and restrictions on moving capital across national borders.

A little background might be in order.

Throughout the '80s and '90s, the IMF encouraged developing countries to liberalize their financial markets and to do it fast. Free market types basically believed that doing this would promote the efficiency of financial markets, channeling funds to where they were needed most and thus leading to better economic growth.

Well, at the beginning of the '90s many developing countries began to liberalize and, due to a combination factors, were faced with massive inflows of foreign capital. To simplify, the US Fed had pushed interest rates down to help the US recover from a recession, prompting investors to look elsewhere for places with higher rates. At the time, Latin America was just such a place. Moreover, apart from having high interest rates, most countries in the region had spent the last decade--the lost one--pursuing neoliberal reforms that investors perceived favorably.

In any case, countries that had been cutoff from international capital markets for years suddenly experienced a huge surge of foreign capital inflows. The problem, however, was that when it comes to capital flows you really can have too much of a good thing. It quickly became apparent that these large flows of foreign capital carried significant risks and posed a challenge to macroeconomic management.

One side effect of capital inflows is that it puts pressure on your currency to appreciate and for many of these countries, which were pursuing an export-led development strategy, this was a big problem.

Many of these countries were also concerned with a loss of monetary policy independence, meaning that the abundance of foreign money was pushing interest rates down despite the wishes of central banks to keep them high in order to fight inflation.

Policy makers were also concerned that these large inflows could make countries more vulnerable to financial crises. In particular, Latin American banks took advantage of all this cheap foreign money and borrowed excessively short-term and in foreign currencies, which put them at risk if the inflows were to stop and the currency to collapse.

The point of this story is that one country in particular managed to juggle all these problems quite well and it did so using capital controls. This country also happens to be the free market darling Chile (Colombia and Brazil have used similar controls with varying degrees of success).

Another famous success story comes from Southeast Asia during the Asian financial crisis. Facing massive capital flight, Malaysia broke with the IMF and imposed temporary controls on capital outflows to prevent it's economy from collapsing. Its economy proceeded to recover rapidly.

Whatever. So now, in spite of mounting evidence and it's own endorsement, the IMF is warning that capital controls could cause significant distortions:

"Since the use of capital controls is advisable only to deal with temporary inflows, in particular those generated by external factors, they can be useful even if their effectiveness diminishes over time... However, the decision to implement capital controls should consider their distortionary effects not just on the individual country, but also on the global economy in the event their use were to become widespread."

To be fair, there's nothing wrong with this statement. Controls on capital inflows shouldn't be used as an excuse to avoid pursuing meaningful macroeconomic adjustments and if controls become widespread they might slow global recovery (a rather large claim backed by little evidence). But it's clear that there are actors within the IMF trying to prevent capital controls from becoming standard policy. So before we rush to the conclusion that the IMF is seriously reconsidering it's long-held policy stances, we should wait to see how it's endorsement of capital controls translates into actual policy.

De jure vs. de facto capital mobility in South America (gratuitously wonky)

I know a lot of you have been waking up in the middle of the night with a cold sweat, anxiously wondering about the state of financial liberalization in South America. What's more, you've got that dreaded feeling of anguish in your stomach, that creeping suspicion that de facto and de jure measures of capital mobility might show different results. Oh the horror! Well, despair no more because maladjusted knows these feelings all too well and is here to put an end to your sleepless nights.

Remember back in the day when the IMF and other international organizations were telling developing countries that they should open up their financial markets? You know, since financial liberalization is basically the same as trade liberalization, it followed that if free trade is super awesome for development so is the free movement of money across borders. I mean, borrowing money is the same as trading goods, only across time, right? Moreover, rich countries have a lot of money and poor countries don't, so letting the rich countries invest all they want in poor countries kinda makes a lot of sense. What could possibly go wrong? It's not like capital flows can suddenly stop, create dangerous asset bubbles, exacerbate the business cycle or lead to financial crises.

Whatever. The point is that developing countries have sought to open up their financial markets to foreign capital and South America is no exception. The graph below takes a first look at capital mobility. We'll call it the "de jure" approach.

[De jure capital mobility, 2008]

This shows how freely money can move in and out of each country in principle. The index was put together using the IMF's Annual Report on Exchange Arrangements and Restrictions (AREAR), which catalogs every member country's regulations and restrictions on international transactions. AREAR distinguishes between 13 different types of regulations on capital flows, ranging from restrictions on financial derivatives to foreign direct investment. In any case, if a country has regulations in all 13 categories it gets a score of zero, meaning that capital is not very mobile. At the other extreme, if a country gets a score of 100 it means it doesn't have any regulations and capital is perfectly mobile.

Chile, Peru and Uruguay have the least restrictions on capital mobility, with equal scores of 84.6, while Argentina, Brazil and Venezuela are on the other end of the spectrum with relatively high restrictions. Now, the outlier here is Colombia, which has by far the most restrictions and a low score of 7.7.

Yes indeed, the US' BFF actually has labyrinth-like restrictions and regulations on international financial transactions. If a type of transaction exists, chances are Colombia regulates it. Colombia pretty much regulates everything from personal capital movements to foreign direct investment (curiously, the only type of transactions it doesn't restrict in some way are those pertaining to real estate).

But what's the problem with the de jure index? Well, it is based on restrictions in principle, which is to say that it only tells us about the laws governing capital flows in each country but doesn't tell us anything about how they are enforced or how restrictive they each are. For example, two countries could both have restrictions on financial derivatives. Say country A has a 2% tax rate on each derivative while country B has a 30% tax rate. Obviously country B is much more restrictive but under the de jure index both regulations are given equal weight. Similarly, A country could have restrictions on almost every type of transaction but in practice might not have the institutional capacity (read: is too corrupt) to enforce them.

This takes me to our second measure of capital mobility. While the de jure measure can be loosely interpreted as the "intention" to regulate or restrict capital flows, the de facto measure below shows capital flows in practice.

[De facto capital mobility, 2003-2006]

This one measures capital mobility in terms of actual flows. It basically takes the absolute value of the capital and financial account of each country's balance of payments (courtesy of the IMF's international financial statistics) and divides it by GDP. In other words, it sums the amount of money entering and leaving a country and shows it relative to the economy's size. Also, in order to avoid the effect of yearly fluctuations and the contraction of flows during the global financial crisis, all the values are taken as an average of 2003 through 2006.

So now that we have both measures of financial liberalization, what can they tell us? The first thing to notice is that there's a bit of a reshuffling in the order of countries, meaning that the countries with the most laws and regulations on capital flows do not necessarily have the least capital mobility. Chile and Uruguay still lead the pack but now Chile dwarfs every other country. Peru goes from the top bracket down to sixth place. Also, Colombia jumps up a few spots.

Another insight is that while Venezuela comes in as the second to last place in the de jure index, it is third according to the de facto measure. The emerging market darling Brazil comes in pretty low in this case, but this is mostly because of the sheer size of its economy (has a much, much larger denominator than any of the other countries).

Well, there we have it folks. It seems that feeling of dread in your gut was justified afterall: de jure and de facto measures of capital mobility paint different pictures about the state of financial liberalization in South America. I hope maladjusted was able to put your mind at ease and help you get some sleep.