The usual hyperbolic claims were made for this tax in advance. It was predicted (by advocates) to reduce consumption by 10 to 13 per cent. They said it would prevent "up to" 630,000 cases of diabetes by 2030.
Obviously, we can't yet measure the effects on diabetes (if any), but can see what effect the tax is having on consumption. Via Jon Fell, I see that the PepsiCo quarterly results have been published. They show a small rise in sales since the start of the year. Regarding their soft drink sales in Latin America in the 12 weeks up to the 6th September 2014, they say this:
Volume increased 1%, which included a contribution of nearly 1 percentage point from certain of our bottler’s brands relating to our new joint venture in Chile. Latin America volume increased 5%, primarily reflecting a mid-single-digit increase in Venezuela and a low-single-digit increase in Mexico, partially offset by a low-single-digit decline in Brazil.
And in the 36 weeks up to the 6th September...
Volume increased 0.5%, which included a slight contribution from certain of our bottler’s brands relating to our new joint venture in Chile. Latin America volume increased 3.5%, reflecting nearly 2 percentage points from certain of our bottler’s brands in Chile, a mid- single-digit increase in Brazil and a slight increase in Mexico.
Coca-Cola will release its latest results later this month, but its quarterly statements for the first half of the year have already been published. In contrast to Pepsi, they show a small decline in sales. In the first quarter of the year:
Volume in our Latin Center (+5%) and South Latin (+2%) business units continued to grow, partially offset by a low single-digit volume decline in Mexico given the new excise tax that impacted the beverage industry and our business.
And in the second quarter:
Latin America’s volume was even in the quarter, as strong 8% volume growth in our Latin Center business unit was offset by a 3% volume decline in Mexico.
So there we have it. Soft drink sales are up slightly for Pepsi and down slightly for Coca-Cola. Coca-Cola is twice as large as Pepsi in terms of sales in Latin America so we can surmise that overall sales are down, but only by about two per cent.
For the anti-soda crowd, this represents success. They deal in a world where there are no costs, only benefits (a defining characteristic of the fanatic) and so any decline in sales is a victory, no matter what economic burdens are placed on the population.
More objectively, however, this is a very feeble outcome from a major tax initiative in a low income country. Once again, the predictions have proven to be hopelessly optimistic. This natural experiment suggests that the elasticity of demand for soda in Mexico is in the region of -0.3, much more inelastic than the computer models assumed. (Soda consumption may continue to decline in Mexico—we shall see—but you would expect the biggest decline to take place in the first months of the tax.)
What effect will a decline of 2 or 3 per cent have on obesity and health? Even if we leave aside the substitution effects that tend to offset any effect on calorie consumption, it is hard to see it being anything other than negligible. Mexico has the highest per capita consumption of soda in the world, but soft drinks still only account for 5 per cent of calories consumed. A low single digit decline in a source of calories that only made up one twentieth of total energy consumption in the first place is so trivial that any impact on weight, let alone diabetes, is going to be too small to measure.
The Mexican soda tax should be seen as what it is: just another stealth tax.