Presentation by the Governor of the BCRA, Federico Sturzenegger, as a guest speaker at the regular session of the National Academy of Economic Sciences.
First of all, I want to thank Juan Carlos de Pablo and the Academy for inviting me to give this presentation today.
I cannot resist the temptation to begin by quoting that well-known work by Juan Carlos, Las desventuras de Julián Falacia (The Misadventures of Julian Falacia), a character who misunderstood economics and always ended up upside down. The first chapter is unforgettable: Julian becomes convinced that the country's economic problem was parasitic intermediation. He then starts a very successful campaign in which supermarkets, transportation, logistics, and packaging were eventually eliminated. Of course, the problem occurred when Julian suddenly came up with buying a piece of cheese. In the story, Julian ends up walking the pampas for days (since stores, transportation, and marketing had been eliminated), looking for a producer who would sell him not a piece, but a whole wheel of cheese (since packaging had also been eliminated). Poor Julian! (This is actually how all the stories ended).
But this reference is not idle nor merely a tribute to Juan Carlos's wit, which would deserve recognition, but it is much deeper and has to do with our tendency to think of problems as partial problems, considering them in isolation and not all the implications and collateral effects they might have. I would like to tell the difference between these two ways of thinking about economic problems: "partial equilibrium" reasoning and "general equilibrium" reasoning. And this problem arises nowhere but in the field of macroeconomics and, particularly, in monetary theory.
Let's warm up with some examples to understand what we are talking about. Let's imagine a price rises. For instance, the energy tariff increases in a given month. In response to this event, it is very plausible that we hear economists make a calculation like this. If the CPI weighting of those rates were 5%, and the increase were 100% (I have made up both figures), one would conclude that the CPI would have an increase of 5% in that month. The reasoning seems perfect. That is the weight in the index, and that is the increase. The calculation yields that result. I would limit this reasoning to the field of partial equilibrium. And the conclusion would end there. The calculation is made, and a conclusion is obtained. I even risk that a consultant who makes this calculation might even charge a fee for presenting it.
Let's take this event to the field of general equilibrium. In this case, the analysis requires further questioning. In the mental exercise we suggest, there is a price that rises. But we also incorporate the budget constraint faced by the family into the analysis. That budget constraint is based on the fact that families have a certain nominal income, which we do not change (it was not changed in the partial equilibrium analysis either), and that is the income families have to spend on all goods. So when a price rises, the budget constraint clearly implies that less money can be spent on other goods, and the price of these goods should fall. If we assume that the demand for real balances does not change, the final result is that final prices remain constant on average. In this case, the economist who thinks in terms of general equilibrium would answer that the effects of the rates on prices would be null.
What was the difference between one reasoning and the other? Simply that one took into account the effect of the higher spending on the goods whose price rose, on the ability to spend on other goods, and on the limit imposed by the budget constraint.
In some sense, partial equilibrium reasoning puts families above their budget constraint, if we considered an inflationary effect. We could assume that the increase in rates increases the speed of money circulation, i.e., it simultaneously generates a lower demand for real balances that allow for a higher level of prices. However, this relationship is rarely made. Or perhaps, the result is that the activity level falls. This argument is used a bit more.
We may look at what happened last month to understand how this simple reasoning helps us think better. In January, inflation in the Autonomous City of Buenos Aires was 4.1%. In February, the government implemented a significant correction in the price of electricity. The first calculations I saw around ventured an impact of 5 to 8 percentage points, mentioning that this would be the additional rise in inflation for that month. However, February’s price index was lower, not higher than that of January. Where was the impact of the rate hike then? Where is it? It was constrained by the very budget constraint. And thus, in February, rates rose sharply and the rest of the goods reduced their increase significantly.
The inconsistency of the partial equilibrium reasoning becomes even more evident if we change the way we pose the problem. Let's imagine that instead of a price increase, we are talking about a tax increase. We rarely think that tax increases are inflationary, precisely because when we think about taxes, we think more about the budget constraint and less about prices. In fact, there is really no difference between the two exercises. The only colleague I have heard making this reasoning, which I consider correct, is Javier González Fraga.
This recognition of the need for equilibrium is what led to the revolution of rational expectations. At first, that approach was too much focused on emphasizing the concept of general equilibrium, with perfectly functioning markets. Later, when they wanted to model market imperfections, which obviously exist, macroeconomics never lost the rigor imposed by the concept of general equilibrium. On this structure, the neo-Keynesian structures were then built. However, macroeconomic analysis models were never allowed not to be of general equilibrium.
However, this academic rigor has been lost in our daily work. And it is present in many of our discussions. Let's take other examples.
A couple of months ago at the Davos meeting, there was a strong discussion about the impact of the oil prices fall on the global economy. Those who argued that the price drop was bad, and were not novices in the task of global economic policy, argued about the loss of purchasing power and the investment will of the exporting countries. In my view, they did not see the receiving countries’ expansive power over the aggregate demand.
A related example that we teach in class all the time goes back to Jeffrey Sachs's contributions in the early 1980s, when he discovered that the increase in oil prices improved Japan's trade balance. A result that was considered counterintuitive at the time. A partial equilibrium economist would have thought that the increase in oil prices for an oil-importing country had to worsen its trade balance, but for a general equilibrium economist, what was relevant was the intertemporal budget constraint. In this context of greater poverty for the Japanese economy—because an input they import becomes more expensive—it is understood that aggregate spending falls and the trade balance must improve.
My favorite example refers to the entire set of policies that are justified because they generate an effect on aggregate demand. Let's look at three examples. The first one has to do with fiscal policy. For instance, it is typically understood that a reduction in spending will generate a drop in aggregate demand, or one might add, this is what a partial equilibrium economist would surely think. But the general equilibrium economist will withhold the conclusion until they understand the impact of lower spending on financing. Does lower spending mean lower taxes? Well then, the drop in public spending must be offset by the increase in private spending generated by the tax cut. As can be seen, the effect is not so linear or simple. Does lower spending mean that the government must stop borrowing in the domestic market? Resources are then freed up for the financial sector to channel them to consumption or production. In this context, reducing the deficit has little to do with aggregate demand, although it will surely imply reallocations in spending.
The fallacy of partial equilibrium also manifests itself in arguments like “public spending is expansive when it transfers resources from high-income segments to low-income segments with a higher tendency for consumption.” We hear this reasoning every day. It fits for a TV program, where the argument appears both correct and sophisticated. “Wait a moment,” a general equilibrium economist might say. What about the resources a person saves? For a moment, imagine that this person deposits them in a bank. Well, that bank would surely lend them out, expanding aggregate demand in another way.
Investment, perhaps? Countries that save a lot are not countries with chronic demand shortages but with high investment rates. If it were true that savings do not generate aggregate demand, China would be the most recessionary country in the world.
Of course, if savings are kept in a can, that demand is affected. But only in that case. And let's note the interesting conclusions this reasoning reaches. In 2015, before lifting the foreign exchange clamp, Argentines desperately tried to accumulate foreign assets. In plain Spanish, buy dollars.
So much so that the wage increases from wage agreements were depressive on aggregate demand. Clearly, a wage increase does not increase aggregate demand; it only redistributes it between companies and workers. As in 2015 workers had a greater incentive to buy dollars and save them than companies did, we should conclude that wage increases in that year were depressive on aggregate demand. But, as far as I remember, no one made this argument at that time.
Another very common example is related to the Procrear program. According to its promoters, this program generates a significant expansion in aggregate demand. However, what is never taken into account is that the use of resources for this purpose is taken away from elsewhere. What is the effect on aggregate demand, and what are the multiplier effects of that expenditure that will no longer occur? For years, the previous government relied on the expansive effect of the aggregate demand of Procrear program (which is also in an industry with many divisions), but the growth in demand was delayed and never materialized. What is increased in one area is taken away in another one.
Thus, a general equilibrium economist is not surprised that years of fiscal improvement mean years of economic expansion, that years of increasing fiscal deficits are years without expansive effects on demand or that, with contractionary monetary policies, the pass-through would not be what it was supposed to be. He would neither be surprised that a rise in certain specific goods did not generate the expected inflationary effect, and so on.
One of the areas where the use of the partial equilibrium concept is most surprising is in monetary theory and inflation predictions. The following conclusions often coexist within the same report. For example, this one, which is correct: price controls are not useful for controlling inflation.
This is so because, as long as there is money to spend and even more if the amount increases, the government can control some specific prices. However, the spending capacity will be expressed in the purchase of other goods, and if certain goods have a controlled price, it will only imply that there will be more spending capacity on other goods and therefore their prices will rise faster. Hence, inflation will persist despite the controls. But as I said, this type of reasoning coexists with other incorrect ones, such as the idea of repressed inflation due to some controlled prices. Believing that inflation is restrained by some prices is believing that controls are useful for controlling inflation.
In general, I have found that it is very common to talk about inflation without referring to monetary market equilibrium. If some prices rise, it is assumed that the general price level will rise. But the price level is determined in a monetary market, which has a money demand and a money supply, and that equilibrium will occur one way or another.
The natural way for such equilibrium to take place is with changes in the price level. Let us assume for a moment that the money demand rises above the supply because the monetary authorities reduce the monetary aggregates and, for various reasons, the prices of goods do not fall. That is, let’s include rigidity in the goods market, which exists for many reasons.
In this case, for monetary equilibrium to occur, either interest rates must rise, the product must fall, or the amount of money must increase. This last case is the easiest to analyze and is known as the passive-money view: prices rise for whatever reason and the policy maker then validates them with issuance. This would be the Barro Gordon model: if expectations rise, then the central banker validates them to avoid a recession.
But let us discard this equilibrium by now, and see what happens where the money supply is also exogenous. This is a more stubborn central banker. Two possibilities then arise: either the interest rate increases, or the GDP level falls, or a combination of both. The fall in GDP occurs because there are no banknotes to buy goods at their current prices, and the interest rate rise occurs because the demand for money pressures the monetary market, generating a tightness that increases rates. These pressures lower inflation, making them compatible with monetary equilibrium.
But often dynamics are analyzed as if price changes followed their own independent path from the markets that define them. The same concern arises in the debate about the pass-through. Many economists take for granted that there is a direct transmission of changes in the foreign currency price to local prices. And obviously, there is a correlation. But this correlation is analyzed without considering the budget constraint or analyzing causality. For that reason, many people have been surprised by the low pass-through of the recent exchange rate unification. But I am not surprised when I include in the equation a very restrictive monetary policy , which imposes greater discipline on prices. This is thanks to the Minister of Finance who has done a superlative job and, in the first quarter of management, has practically reduced BCRA's requests for financial assistance to zero. Could the pass-through be discussed without conditioning it on monetary policy? I do not think so. But in Argentina, it is discussed as if it could.
In fact, let me conclude with an analysis of the current situation in terms of prices and rates.
There is obviously a view that inflation is high, and it certainly is. This would be one possible reading. But let me see it from another perspective. We started from a steady state last year with a 2% inflation rate. Surely, in the first four months of the new government, a total of about 8 more points of inflation will be added to the existing price trend. Let me then share a second reading. My view emphasizes that, over the last few months, many prices have been readjusted, including the exchange rate movement that reached almost 60% (now a bit less); taxes (read: export duties) and trade restrictions were reduced, having an impact on the price of many tradable activities, and tariffs were significantly adjusted, reversing years of price and investment backlog. Considering that it all occurred at the cost of only 8 inflation points is worthy of a positive reading on the situation. Not an amenable reading, but a positive one.
How was this result achieved? Through a very prudent monetary policy that we have implemented since December 10. We can analyze it from many angles, but monetary equilibrium is what will ultimately determine inflation.
Recently, at a BIS conference, I heard Mario Draghi recall the following quote: “Similarly, Fed Chairman William Miller observed in his first FOMC meeting in March 1978 that, ‘inflation is going to be left to the Federal Reserve, and that’s going to be bad news. An effective program to reduce the rate of inflation has to extend beyond monetary policy and needs to be complemented by programs designed to enhance competition and to correct structural problems’. It was only when Paul Volcker arrived as Chairman in 1979 and shortened the policy horizon that the Fed took ownership for controlling inflation. Inflation, which peaked at around 15% in March 1980, fell below 3% by 1983.”
Our models indicate that monetary policy has a lag of about six months, meaning that the inflation we are seeing is the result of a very strong expansion in the monetary aggregates that occurred in the second half of 2015, with effects that will continue to unfold during the first half of the year. Added to this was the liability generated by the irresponsible sale of dollars for future delivery during that brief period, which was sterilized during our administration, at the cost of an increase in the quasi-fiscal deficit which, although we could manage it, still imposes a cost on all Argentines.
To conclude, I go back to the beginning: thinking about the economy from the perspective of general equilibrium has great advantages and serves as an excellent guide for economic policy. Never leave home without fully considering that framework.
Thank you.
