Inflation has memory | overhead

Hi how are you? Although we are four days away from the start of the World Cup and that mobilizes the whole of society, the phrase of the Minister of Labor, Kelly Olmos, generated -logically- a wave of criticism that made her, after a few hours, have to go out and apologize.

The criticisms focused more than anything on the fact of comparing the inflationary escalation, a very serious problem that affects the entire population (and, above all, the lower income sectors), with a World Cup. But the first part of his sentence stuck in my head: “I believe that you have to work all the time because of inflation, but a month is not going to make a big difference. On the other hand, from the emotional point of view, of what it means, we Argentines want Argentina to win.”

Is it true that a month doesn’t make a big difference?

Suppose that the World Cup impacts inflation in November, making it one percentage point (pp) higher than it would have been without the World Cup (actually, it would affect mid-November and mid-December, but to simplify the bills). In addition, suppose that in December there is a 0.3 pp reduction due to the application of the “Fair Prices” program, in line with the expectations of the government.

As you can see in the table below, the contribution that November’s poor performance would have on annual inflation would be quite moderate (102% vs. 98% annually), in line with what was proposed by the minister.

Do you know how Cenital is financed? The main support of our medium is its readers. That makes us proud and gives us the necessary peace of mind to do the journalism we believe in. If you like what we do, Join our community too.

Inflation projection with and without World

Source: self made

However, the panorama changes if we extend the analysis horizon. Now let’s see what would happen during 2023. To do this, and continuing with the government’s perspective, let’s assume that inflation is reduced in line with what is proposed by the projections that appear in the 2023 Budget. In this case, what we do to avoid comparing ” pears with apples” is to apply the same rate of inflation reduction in both cases, so that in the “without World Cup” scenario, annual inflation for December 2023 would be 60% (as it appears in the budget). But look what happens now with the World Cup setting. As the graph below shows, in this case inflation for the end of the year would end up being 13 pp above the other scenario.

Inflation projection with and without World

Source: self made.

As we said, in both cases the same disinflation rate was applied, with which the gap of 13 pp is explained only by that percentage point of difference that took place in November. The explanation is strictly mathematical: if you start with large values ​​and apply the same rate, the small differences become larger and larger over time.

If, for example, instead of starting with annual inflation of 102% and 98%, it had been 10.2% and 9.8% (that is, exactly the same, but removing a zero), for the purpose of year inflation would have been 6.1% and 7.4%, respectively. The difference doesn’t seem that big there, does it?

a non-reversible process

But, apart from the purely mathematical question, there is an additional element that I think is relevant to highlight. inflation has memory. This means that, as time goes by, the behavior of people in an economy with (high) inflation changes, because they learning of your mistakes.

For these reasons, we could define memory as the ability of inflation, at a given moment, to preserve certain characteristics, despite the fact that the causes that originated it are no longer present at that moment. This means that the behavior of this phenomenon cannot be explained only by current circumstances, but by the entire process accumulated over time.

In more technical terms, what we are saying is that inflation is a process.”not reversible”, that is, when subjected to the inverse circumstances that explain its current behavior, the process does not return to its previous state. If people modify their behavior based on learning, it is very difficult to make them “unlearn” what they have learned.

For the specific case of inflation, there are various behaviors that change over time. For example, the way in which companies set their prices, our savings decisions, the mechanism and the frequency of parities, etc. All these issues have been changing over the years (and the increase in inflation), making the structure -and, therefore, the behavior- of the current economy not the same as it was 10 years ago. This means that, for example, a 50% devaluation today would not have the same impact on prices as it would have had 10 years ago (clearly today it would be much greater).

In order not to extend ourselves too much (and because with an example it is enough to see where we are aiming), let us analyze the most relevant case, the fixing of prices by companies.

Sometimes something transitory becomes permanent

Broadly speaking, companies set their prices taking into account their costs and the prices of the competition. But to do this, they not only take into account their current situation (and that of the competition), but they must also have an estimate of what they think is going to happen in the future. The more accurate that projection is, the better your performance will be.

Let’s see an example. Let’s imagine that in a product there are only two companies that manufacture it (1 and 2). Both sell at the same price and have the same market share. At a certain moment, costs increase suddenly, caused -for example- by the outbreak of a war (what we call a “shock”, which unbalances the initial situation).

Firm 1 believes -incorrectly- that the war will last for several years, so it decides to increase its price in line with the increase in costs (so as to keep its profit margin constant). On the other hand, company 2 believes that it will last very little, and therefore decides to keep the price unchanged (it reduces its profit margin, absorbing the increase in costs).

This leads to company 2 increasing its sales at the expense of company 1 (which now sells at a higher price) and, very probably, to its profits being also higher than those of company 1 (this will depend on the elasticity of demand for the product).

After a few months, when the war ends, the companies rearrange their prices, which are again identical. But, nothing ensures that the sales of the companies will be the same as before. It is likely that some of the people who happened to buy from company 2 realize that that product is better or there is something they liked better. If so, company 1 would lose customers significantly. permanent against company 2. In other words, the disappearance of the shock transient it did not return the situation to its original state.

This example, in addition to showing a non-reversible process, shows how important it is for companies to make estimates that are as accurate as possible. The issue is that information is expensive, and not all companies can afford to pay an inexpensive consultancy (later we would have to discuss how much better that consultancy does its projectionsbut that is for another edition).

And it’s not just a matter of estimating correctly, but also of estimating in relation to the other company. Notice that, in the previous example, if both companies had estimated the same thing (it doesn’t matter if they got the duration of the war right or not), the process would have been reversible (if both companies estimated correctly, they lowered the price, but continued selling the same thing, and if they did it incorrectly, both would raise the price, but they would continue selling the same thing).

The combination of these two elements (the fact that it is expensive for companies to collect the information necessary to estimate correctly and that what the rest does is important) leads us to the BCRA, because this institution -as it is in charge of controlling inflation – becomes an automatic reference -and free- for companies.

Of course, for that to happen, the BCRA has to do its homework well. The most interesting thing about all this is that, if it did so, a virtuous process would take place where, as companies see that the monetary authority estimates inflation and complies, they pay more and more attention to what it says and use said projection as your own reference.

If we take it to the extreme (and unreal) case, note that the BCRA could control inflation just by announcing its estimate, making companies automatically adjust their prices based on that.


This serves as a point of comparison to see what happens when the BCRA does not do its job of controlling inflation well (or announces a projection that is far from reality, it sounds similar, but it is not the same). In this case, the exact opposite happens and it becomes a vicious circle: the lower credibility of the BCRA makes it increasingly difficult to control inflation and, therefore, make it align with its projection (removing even more credibility).

As this happens, inflation not only increases, but also becomes more erratic and difficult to estimate (for the BCRA or for anyone who wants to). In particular, this leads companies to focus more and more on their costs -the variable they can accurately monitor- and to transfer all kinds of “shocks” to prices (since they have no way of knowing the duration of the shock). ) to hedge against future losses.

What is the conclusion of all this? What we analyze At the time, the Argentine economy a few months ago entered what is known as a high inflation regime. One of the characteristics of this phenomenon is precisely that companies do not have a specific reference when setting their prices, that is, there is nothing that works as anchor, which means that, given the slightest shock, prices increase and last over time. With which, yes, one month can end up making a significant difference, because small variations accumulated over time can cause big changes.

Bonus track

  • Now in a little while, at 11:30 am, there is a panel from the annual meeting of the Argentine Association of Political Economy (AAEP) that fits exactly with these themes: “Stability, expectations and price dynamics”, in which four specialists will speak about it. can be seen online here.
  • Today I am not going to recommend a podcast. I have something much better. “Estéreo” is the first podcast festival to be held in Argentina. It will take place this Thursday, Friday and Saturday in the city of Rosario. The activities that will be carried out will bring together the most important podcast producers in the country and in Latin America. All the information, here.

This was all for today. If you can’t wait until next week, we’ll follow her for here. And if you liked the newsletter, I invite you to collaborate with Cenital, because the contribution of our readers is our main income.

I send you a big hug.


Inflation has memory | overhead