Yada, yada, yada

An argument that has been floated recently, to the point of meriting the dubious honor of being officially dismissed in the Copom minutes, is that the increase in foodstuff prices would have an effect on Brazilian consumption similar to the increase in oil prices in US consumption, namely that it would work as a “tax” on consumers, reducing demand, hence inflation. Whereas, as I said, committee members downplayed (correctly) the argument, the minutes did not bother to explain why the argument does not make much sense (and, mind you, I am attempting to be polite here). In light of that, I believe that going into more detail might be helpful.

For a start, I think the argument gets dangerously close to circular self-contradicting reasoning. After all, if the rise in foodstuff prices reduces real income, therefore consumption, and then inflation, what would prevent a rise in all prices (aka “inflation”) from reducing income, hence consumption, and – at the end of line – inflation itself. By the same token, a reduction in inflation would (yada, yada, yada) increase inflation, which should already make anyone a bit suspicious of the idea.

That said, trying to avoid this sort of game, we can have a more serious look at the argument only to make sure that it can be indeed safely dismissed. One way to look at it is to distinguish between supply and demand shocks, since the first implies indeed a reduction of real income, whereas the second does not.

Indeed, consider first the case of a negative terms of trade shock, that is, the increase in the price of imported goods, such as the fuel prices for the American consumer. In such scenario the country as a whole transfers part of its income to the rest of the world, with potentially negative impacts on consumption, whose magnitude should depend on the perception of how long this should shock should last (a permanent shock have larger impact on consumption than a transitory shock).

Consider now a domestic supply shock, say, a sharp decline in domestic food production due to bad weather or a plague, which drives up foodstuff prices. This would reduce non-food producers income (since they now have to spend a higher share of their budget to acquire the same amount of food), and would most likely reduce producers income as well. In this case too, the rise in foodstuff prices would also work as a “tax”, and reduce demand.

That said, the faithful reader might recall a note I published recently (The Untamed Lion, August 30, 2007) in which I examined the nature of the rise in foodstuff prices from the Brazilian perspective to conclude that it was stronger demand (both external and domestic), rather than a shortage in local supply, that was driving up food prices. In this case one cannot talk meaningfully of higher food prices acting as a tax, since the non-producers loss is a producer gain.

In the case of a closed economy, in particular, non-producers loss is precisely equal to producers gain and then one would have to elaborate further the distributive effects to figure out the final impact on consumption, that is, which group (producers or non-producers) spends more in consumption. In the case of a small open economy, however, a positive terms of trade shock increases income, that is, non-producers losses are unequivocally smaller than producers gains. Does this ring any bell?

Summing up, the notion that higher foodstuff prices would reduce domestic demand does not find strong support in economic theory, if the origin of higher prices is demand itself, rather than a supply shock. Moreover, evidence on the performance of demand at the margin also does not lend support to the notion that foodstuff prices would cause demand to decelerate. Yada, yada, yada, the Copom was right in dismissing this argument as truly irrelevant.