Argentina’s Bogus Miracle

It is time to debunk the myth that Argentina performed an economic miracle by actively managing the real exchange rate (RER) and to demand the authorities in that country that they implement a serious economic plan before conditions further deteriorate. The myth is that Argentina’s strong post-crisis recovery was due, primarily, to the maintenance of a high and stable RER supported by government intervention and a prudent mix of fiscal and monetary policies. In reality, the high rates of growth observed since 2Q02 were due, primarily, to favorable international conditions and, secondarily, to an undervalued (hence, unstable) RER. Fiscal policy was expansionary and monetary policy was, at best, accommodating.

Let us begin by examining the facts. When the last financial crisis erupted in Argentina in 2H01, the RER was, clearly, overvalued. Due to a sudden stop in capital inflows, low terms of trade, a strong USD, and a depreciated BRL, the RER level needed to ensure full capacity utilization and external balance was about 25% lower than the actual one. But, since the Convertibility law prevented nominal depreciation, a quick adjustment was not possible and, as a result, the output gap and the current account deficit were unsustainably high. Argentina reacted to the crisis in 1H02 by defaulting on the public debt, then devaluing the peso. Relative to December 2001, the ARS is, even today, 50% cheaper in real effective terms. Yet, soon after these measures were introduced in 1H02, international conditions improved dramatically. Export prices increased, world interest rates declined, and the government was able to restructure the public debt in very favorable conditions. Moreover, the USD depreciated vis-à-vis the other main currencies of the world while the BRL appreciated in real effective terms. The combination of these effects was a major increase in disposable income, real GDP, the RER, the current account, and (since much of the external windfall was taxed by the government) also the federal and provincial budgets.

Thus, in retrospect, the 50% depreciation of the RER was, probably, not necessary. Of course, the government that lived through the crisis in 1H02 did not know that, a few months later, international conditions were going to improve as much as they did. The question is why, once that became clear, neither they nor the administration that followed let the RER appreciate but, to the contrary, made RER undervaluation an explicit (and key) policy objective. To some analysts, the reason is pretty obvious: by increasing export competitiveness, the low RER stimulates growth. The problem with this argument is that it does not differentiate between a low equilibrium RER and a low disequilibrium one: one thing is to reduce the RER by, for example, reducing government spending as as share of GDP and another is to do it by repressing inflation. The first policy generally leads to a sustainable increase in growth. The second does not (although it may exacerbate aggregate demand in the short run).

How did the government of Argentina manage to undervalue the RER? Very simple: it instructed the central bank to intervene in the FX market (to prevent nominal appreciation) and sterilize part of the excess supply of money resulting from this intervention (to limit real appreciation). Since this was not enough to do the job, the government also increased export taxes on food and fuel and imposed direct controls on basic consumption prices, formal wages, and the tariffs of most energy products and public services (incidentally, this is not very different from what China is doing). Thus, RER undervaluation caused a second round of expansion that drove the economy well beyond the point of adequate utilization of the installed capacity causing inflation to accelerate much more than the official records show.

Maintaining the RER at its current level without adjusting the fiscal/trade/incomes policy mix, will exacerbate distortions and, ultimately, become unsustainable. This, assuming the international environment remains favorable. If, on the other hand, the latter becomes unfavorable, Argentina’s twin surpluses (external and fiscal) will vanish and the RER will probably have to depreciate even more. The risk is that, in this context, the government will lose control of inflation. If this happens, the economy will, most likely, stagnate. Consider the following two alternative scenarios.

First, suppose, optimistically, that the world can avoid a global recession. To ensure adequate economic performance in the long run, Argentina’s growth model must undergo fundamental changes. For starters, domestic prices must be liberalized. This includes the realignment of tariffs in the energy and public services areas and the elimination of price controls on basic consumption items. These policies would inevitably produce inflation acceleration and a decline in the RER. The only way in which the government can ameliorate these effects is by implementing a tighter fiscal policy accompanied by trade (particularly, import) liberalization. But, this is opposite to what Argentina has been doing in the last three years.

Now, let us imagine there is a global recession. If this happens, Argentina will be ill-prepared to absorb the external shock . Unlike Brazil, which let the real exchange rate appreciate and the overall fiscal deficit to contract during the expansionary phase of the cycle, Argentina applied the opposite recipe: it let the real exchange rate depreciate until the latter became grossly undervalued, and allowed the fiscal surplus to fall in spite of record-high growth in tax collection (see graph). By implementing this highly pro-cyclical mix of policies, Argentina did exactly the opposite of what the doctor ordered. So, now, when rational policymaking would dictate that the currency be let depreciate in response to weaker external fundamentals and the fiscal surplus fall in response to an endogenous reduction in tax collection, there is little room for both.


Addendum: The case for a depreciated RER

Export-led growth—the idea that preserving the profitability of the export sector helps improves growth performance—has many adherents not only among professional economists, but also among politicians. Yet, not every advocate of export-led growth really knows what he/she is talking about. For some, it is merely a question of avoiding the “foreign exchange gap,” whatever that means. For others, it is the fact that productivity is higher in the exportables sector.

The problem with these linear interpretations is that they can be used to justify undervaluing the domestic currency as a way to jumpstart a growth process. Consider, for example, the argument that, because labor productivity grows faster in the tradables sector, the government should keep the RER as high as possible during takeoffs. This is like putting the “Balassa-Samuelson effect” on its head. What Samuelson and Balassa showed was that the difference in labor productivty growth observed in the tradables and nontradables sectors explains why the RER appreciates more in fast-growing countries than in slow-growing ones. If there is a causality, it goes from growth to the RER rather than the other way around. Moreover, the correlation is negative rather than positive!

While it is commonplace to invoke the role of a depreciated RER in the takeoffs of sucessful East-Asian exporters—Japan in the fifties, Korea in the sixties and seventies, and China in the eighties, nineties, and even today—the link between both was more sophisticated than RER activists typically assume. In those countries, income growth was reinforced by a phenomenal increase in saving and investment rates. Typically, saving rose more than investment, causing large trade surpluses and RER depreciation. Notice, however, that the likely cause of growth was not the high RER itself, but the fact that, to get to it, countries had to increase national income faster than consumption.

18 Responses to "Argentina’s Bogus Miracle"

  1. Anonymous   September 21, 2007 at 8:19 am

    Excellent critique of causes and currency policy role in the recent Argentina economic “miracle”…Well said!

  2. Vitoria Saddi   September 21, 2007 at 9:01 am

    Joaquin, This is perhaps one of the best analysis I have read on Argentina this year. I could not agree more with the points you raised. In a sense, the ongoing debate in Argentina reminds me of the 1990s and the convertibility plan. In other words, like then, now we have people like you who is (correctly) against the current set of policies. On the other side, you have Eugenio Dias Bonilla (see his latest blog on growth in Argentina) who is clearly defending this strategy. Why two great economists can agree to disagree?

  3. Alejandro   September 21, 2007 at 3:17 pm

    Joaquin, I have a few comments. I hope you find them useful. First, you said “…In reality, the high rates of growth observed since 2Q02 were due, primarily, to favorable international conditions and, secondarily, to an undervalued (hence, unstable) RER…” I disagree. There’s no question that the international environment helped but you are overstating its relative importance. If you check the numbers carefully, the recovery (started in Q3 2002) was led by the increase in the domestic demand components (both, consumption and investment expenditure account for 85% to 120% of the increase in activity!). For the most part of the recovery, we observed that the net exports were indeed a contractionary force (except in the very beginning). So, external conditions are crucial in a country like Argentina but we can’t exaggerate their importance. Regarding the relevance of the RER throughout the whole process, I think you should read the literature (for instance Rajan et al, Rodrik, and Frenkel and Taylor among others) that supports the view that undervaluation stimulates economic growth. There are three channels by which the real exchange rate affects employment: macroeconomic, development and labour intensity. The preservation of a stable and competitive real exchange rate plays a crucial role in the promotion of employment and economic growth (Not enough space to elaborate here). Having said that, I do think that there are several mistakes in the policy mix that the government has been using, particularly, an expansionary fiscal policy; they should have run a tighter one and possibly, counter cyclical. Also the lack of an answer for the energy shortage is alarming. I don’t know if we can call it a miracle but you can’t deny that NOBODY expected back in 2002 a recovery like the one we observed even with a benign international landscape.

  4. Mick Rolland   September 22, 2007 at 1:04 pm

    Very insightful post by Mr. Cottani and great comment by Alejandro.   It is indeed true that nobody expected a GDP growth performance in Argentina since 2002 like the one that has actually happened. I would add some factors that I believe are important:   – Convergence to potential production. After the cataclysmic 2001-02 crisis, the reduction in uncertainty (which is usually massive before an expected devaluation) contributed to a natural recovery.  – Monetary and fiscal stimulus in an environment of surprisingly high credibility (possibly attributable to strong fiscal revenues and the favourable international environment) and low interest rates.  – Energy policies (particularly price controls) that provided a tremendous stimulus to other sectors -energy almost for free.  Those two last factors are a clear case of ‘dynamic inconsistency’: The short-run effects are positive, but long-run effects are negative, with an overall negative impact due to the distortions introduced and the likelihood of disorderly adjustments.   The government, as a result, faces now a very tough situation in the energy sector: If it raises prices, it will become hugely unpopular -with powerful and aggressive unions (the “piqueteros”), revolts are probable. On the other hand, if it doesn’t raise real energy prices, it will have to hand out massive subsidies to the energy sector and face a fiscal and debt crisis.   There are two ways out: an immediate lifting of price controls and a consequent fall of the government (due to high political costs and substantial economic adjustment costs) or a gradual adjustment of managed prices -fiscally painful (due to the necessary subsidies)-, that will surely be unpopular anyway. The alternative is to let the energy sector gradually bleed to death until there is no option but its nationalization and subsequent massive subsidization. The implicit costs: a high risk of a major energy shortage, also politically dangerous.   Regarding the real exchange rate, I agree that it should be competitive, but I believe Mr. Cottani’s explanation of the sustainable sources of competitiveness addresses this issue: The key is a favourable environment for savings and productivity growth. Using monetary policy to undervaluate the exchange rate is risky -the government and Central Bank then depend on their fragile credibility.

  5. mark turner   September 23, 2007 at 9:41 am

    Joaquin,  The Argentine auto industry has blossomed and continues to grow and receive foreign investment under the weak Peso regime. The Argentine auto manufacturers council are strong proponents of weak Peso and are actively lobbying to keep the current policy in place. The sector has been a impressive success story in the last 5 years and has little to do with external factors such as commodity prices and the decadence of the US dollar.  Why is this body of businessmen with a strong track record over the last 5 years still against any govt policy to allow the ARS to revalue? Do you think they are ignorant of the full macroeconomic picture in Argentina?

  6. Guest   September 24, 2007 at 10:02 am

    Professor Cottani, Sorry if my doubt is nonsense but I have difficulties trying to grasp this phrase “…Typically, saving rose more than investment, causing large trade surpluses and RER depreciation…”. I thought that when big trade surpluses arise, the consequence you could expect is an appreciation rather than a depreciation of the currency. Am I wrong? Apart from that, great post! Alejandro C. García Cintado

  7. Mick Rolland   September 24, 2007 at 10:45 am

    I will try to reply to Alejandro’s doubt:   A current account surplus may be an indicator of excess savings and thus pressure for an appreciation of the exchange rate. That is a normal case (e.g. China) and you got it right.   However, professor Cottani addresses the possible causes of this current account surplus. This surplus may have nominal causes (the exchange rate) or real causes: real competitiveness (productivity, quality of the production, or in general high value of what is produced and tradable).   We get to the point: when a country increases its real productivity (let’s say because of a climate more favourable to doing business, for example), it translates into lower Unit Labor Costs (ULCs) by definition. As ULCs is usually the relevant deflator in calculating the REER, a decrease in ULC, ceteris paribus, amounts to a real depreciation, again by definition.   Seen from a savings-investment perspective, when there is an increase in production and there is a preference or incentive to save, there may be an increase in production not matched by domestic demand (thus an increase in savings).   The subsequent current account surplus may then tend to exert upwards pressure on the nominal exchange rate. This may be an equilibrating mechanism (à la Balassa-Samuelson), but is probably a consequence of previous real factors in the case of China.  As far as I can get it, this is relevant as Mr. Cottani, in my opinion very lucidly, points out that most economic cause-consequence relationships are interpreted the wrong way in most economic models and policy frameworks.

  8. Joaquin   September 24, 2007 at 11:00 am

    Alejandro,  It all depends on what part of the exchange rate cycle you are thinking of. First, let’s be clear I’m talking about the real exchange rate, not the nominal one. Suppose the nominal exchange rate is growing at 20% per year and so is domestic inflation (foreign inflation is zero, hence the RER is constant. The trade account is balanced. Now, there is positive supply shock due, for example, to the incorporation of thousands of previously underemployed workers who now become part of the productive economy. The effect is an increase in the supply of all goods, tradables and nontradables. The fact that national income increases more than consumption and investment implies that the trade account has improved. Moreover, the increase in the supply of nontradables implies that domestic inflation must have decreased. Given the same rate of nominal depreciation, the RER therefore improves. If this had been a once-and-for-all supply side shock, the dynamic adjustment process would then call for a RER appreciation of the kind you have in mind. This is because, over time, the trade surplus would raise the supply of foreign assets, causing the nominal rate of depreciation to fall, while the increase in national wealth would raise absorption causing domestic inflation to rise (and the trade account to fall). However, if what we observe is a series of supply-side shocks, like the one referred above (think of China), then the effect would be sustained RER depreciation before any meaningful appreciation actually occurs.

  9. Joaquin Cottani
    Joaquin   September 24, 2007 at 2:33 pm

    Very interesting comments.  Vitoria: you are very kind. Not even my mother ever called me a “great economist” (not that she should have either).  Mark: the automobile industry is a clear example of a sector that is doing well because of misaligned relative prices. A growth champion, it is not. Maybe the fact that, in Argentina, cars are a good hedge against inflation has also helped.  Mick: Thanks for your explanation. It is better than my own.  Alejandro: My point is that the series of favorable shocks that Argentina enjoyed since 2H02 (and which, by the way, go well beyond the price of soybeans) eliminated the overvaluation gap in the RER that existed in 2001. By implication, the devaluation and subsequent effort of the government to prevent real appreciation via price and wage controls resulted in real exchange rate undervaluation. Let me clarify: I have nothing against a competitive and stable RER as long as it is supported by good fundamentals, including tight fiscal, trade liberalizaton, and even, if practicable, sterilized intervention (tight money) Where I draw the line (and, I suppose, also you, Rodrick, and Rajan) is with using direct wage and price controls to gain competitiveness by repressing inflation. Finally, I don’t think the effect of external conditions can be measured by the contribution of net exports to GDP growth. You say domestic absorption increased. Of course it did, since the external windfall was equivalent to an huge increase in disposable income.

  10. Rafael Loring   September 25, 2007 at 4:05 am

    Great comments. One of the most lucid analysis I have read on Argentina.  I’d like to follow up on Joaquin’s comments on the external contribution to the economy by adding that the external surplus and reserve accumulation permitted the expansion of internal demand without the problem of having to resort to international capital markets to finance it. Argentina will have to face this problems in the following years as the CC surplus diminishes and the production meets its limits on many sectors suffering from underinvestment.  Rafael Loring

  11. mark turner   September 25, 2007 at 6:57 am

    It beats me as to why you think the auto industry isn’t a growth champion.

  12. Mick Rolland   September 25, 2007 at 11:50 am

    My two cents on the last question:  There are several possibilities of sectors that flourish at the expense of the rest of the economy. Although it may seem a rare case, it is quite frequent in economies with high degrees of protectionism or misaligned exchange rates.   Jagdish Bhagwati, in one of his “immiserizing growth” models, explained masterfully this possibility. If there are significant subsidies or tariffs (which constitute an implicit subsidy at the expense of consumers and other sectors) and the protected sector grows, the economy might be even worse off than with the sector underdeveloped.   The key to understanding this case is the concept of opportunity cost: Without the distortions (tariffs, subsidies or a misaligned exchange rate), other producers (more efficient) mught have enjoyed investment and prices in the protected industry been lower for consumers (thanks to imports). For this to happen, of course, labour and capital markets have to be somewhat flexible.   In the end, the action of the Central Bank that undervalues the exchange rate might be equivalent to a subsidy on certain exporters and a tax on importers and sectors hurt by inflation.   I don’t know the precise situation of the auto sector in Argentina. It may be a case of “immiserizing growth” or not. If tariffs and red tape for imports are high in the sector, or we suppose a misaligned exchange rate (which might not be the case), there is some probability that it is. However, we cannot know it for sure until there is trade liberalization or movements in the exchange rate. (If the sector sees substantial adjustment then, such a diagnosis would seem founded.)

  13. Guest   September 25, 2007 at 1:03 pm

    Many thanks to professor Cottani and the other discussants for the crystal clear explanation they made. I really enjoyed the post and the comments and also I think I learned some economics in the process. Best Alejandro C. García Cintado

  14. Max   September 25, 2007 at 5:43 pm

    Here’s a miracle. In the 26 years between 1976-2001 trade in goods and services has averaged 17% of GDP, while in the 5 years in the 2002-2006 period it has averaged 43%.  So one could say this is by far the most open economy Argentina has had in over 30 years.

  15. Joaquin   September 26, 2007 at 12:53 pm

    Max: The figures you quote are at current prices. At constant prices of exports and imports, Argentina is as open today as it was in 1998. That’s no miracle. Let me explain. Argentina became a more open economy in the early 1990s. By 1998, the share of X+M in GDP was 25%. Today, it is 26%. It is only after you multiply this ratio by an undervalued exchange rate (overvalued tradables) and record international export prices that you get 44%. My guess is that, if Argentina had contined to implement market oriented policies, today the rate of openness (measured at constant prices) would be a lot higher than 26%.  Mark: An industry that has existed for half a century and is only capable to export to Brazil thanks to the existence of a special trade regime within Mercosur and an extremely favorable exchange rate is not a world champion. Fifty years later, the infant industry argument for protecting it no longer applies.

  16. Joaquin   September 26, 2007 at 2:24 pm

    I’m afraid I may have confused some readers with my use of the word “undervaluation.” In my definition, a country’s exchange rate is undervalued if the relative price of nontradables to tradables is too low for the nontradables market to clear. This can only occur, in the short run, if prices are “sticky” and, in the longer run, if the government interferes with domestic price flexibility. I’m making this clarification because I noticed that, in the context of the US-China debate on the exchange rate, many people who claim that the RMB is undervalued say so because China controls capital outflows, intervenes in FX markets, and sterilizes the excess monetary expansion resulting from FX intervention. In my book, this does not necessarily imply undervaluation. Actually, I believe that this strategy, which is typically referred to as the “Asian growth model,” is perfectly legitimate and consistent with free markets. And, while I’m not a fan of sterilized intervention, I think that, in some cases (i.e., when domestic interest rates are sufficiently low), it may work. In particular, sterilized intervention works when the fiscal surplus is high since this amounts to a cost-free way of mopping up liquidity, hence relieving the central bank from having to do itself by issuing sterilization notes and bills, hence preventing inflation to accelerate, hence helping the RER to remain competitive.   Having said so, I do believe the RMB is undervalued. But, not for the reason other fellows would tell you. I think that the RMB is undervalued in real effective terms because, as in Argentina, domestic prices are far from flexible. In other words, they are, to a large extent, controlled by the state.

  17. mark turner   September 27, 2007 at 9:13 am

    Joaquin,  You seem to miss the point.  The term you used was ‘growth champion’, not ‘world champion’. I would certainly agree that the Argentine auto industry’s recent growth could not put it into the category of Korea, Japan etc.   So returning to the phrase in question, would you not say that the Argentine auto industry has been a champion of growth in the last 5 years, which after all is the critical period that needs to be examined?   You also seem to contradict yourself. One the one hand you say that the FX policy used by the Argentine government has not helped growth. On the other hand you say that the auto industry has benefited from ‘an extremely beneficial exchange rate’. Correct me if I am wrong.  Finally, I believe that Mercosur, with all its rules and regulations, existed long before the auto industry’s renaissance. I don’t see anything wrong in taking advantage of favourable trade conditions. In fact, it seems to have been well planned!

  18. Joaquin   September 27, 2007 at 11:38 am

    Mark: (Sorry. I meant ‘growth champion’; ‘world champion’ was probably a soccer mental lapsus, since I was thinking of Brazil and Argentina). I think Mick Rolland already expressed my feelings about why an overprotected sector cannot be considered a growth champion even if, for a certain period, it actually grows. By this token, the import substitution model Latin America implemented in the 50s, 60s, and 70s was successful since it also helped many protected industries to grow. But, as we know, this happened at the expense of the unprotected ones, causing high overall growth volatility (stop-go cycles), very high inflation and, eventually, lasting stagnation in what became known as the ‘lost decade’ (the 80s). In sum, it all depends on what you mean by ‘growth.’  With respect to Mercosur, I’m not saying it was a bad idea, just that the most important exception to free trade within the area is the “special regime” created for the auto industry, that practically forces Brazilians to buy cars and parts from Argentina, and vice versa. Sure the regime also existed in the 1990s. It was as bad then as it is now. With respect to the bilateral exchange rate, I didn’t say ‘beneficial’ (as in good for the country), but ‘favorable’ (as in good for the auto industry, namely, an implicit subsidy paid by other sectors).