I had an interesting dinner with two macroeconomists, Paulo Lins and Michael Navarrette. A follow-up conversation led to me skimming this paper, which dives into the regional heterogeneity of food inflation. Now inflation is not something we typically think of having particularly local or granular heterogeneity (“inflation is everywhere and always a monetary phenomena”, etc, etc), but it’s important to remember that the biggest difference between chalkboard inflation and real-life inflation is measurement.

Economic data is something we often take for granted, in no small part because it’s the substrate from which so much economic research is grown. To fight over it almost feels like nihlism. But just because we aren’t fighting about it doesn’t mean that measurement is easy. It is, in fact, brutally challenging for a host of reasons. Now, a lot of those reasons come down to the demand for immediacy in measurements, which can in turn be dealt with through updates over time. But there’s a deeper challenge that we shouldn’t lose sight of.

Every good is a bundle.

A tomato is a vegetable that is secretly a fruit. Sometimes the price is higher, sometimes it’s lower. But here’s the rub: sometimes when the price is higher it’s secretly lower, and vice versa. Sometimes it’s the same, only it’s not. Sometimes that modestly increased price is secretly a catastrophic increase threatening marinara all across the nation.

Yesterday the tomatoes I bought were 3 for $2. Today they were 2 for $1.50. A modest 11% increase in price. Ah, but see, it isn’t.

The tomatoes today are a little smaller. They came from farther away, representing a seed line that is more tolerate of travel and refrigeration. They are less uniform in color, more acidic, less sweet. Diving deeper, we find that the cost for a 100 lbs of tomatoes purchased in bulk were unchanged. There was, however, less variety to be chosen from because those crates of bulk tomatoes were increasingly curated to fit the needs of Sysco, the chief purveyer for mid to lower tier restaurants, which needs them more for median-customer approved red sauces than spinach salads and bruschetta.

So, dear reader, I ask you – did the price of tomatoes go up? For me, they certainly did. For the median American they barely budged. For Pizza Hut they may have actually gone down!

It’s easy to see how complex goods are bundles of attributes, but it’s amazing how products as commodified as sand or amino acids for livestock feed can quickly become bundles once you put yourself in the shoes of the customers for those goods. When quality, timing, and uniformity enter the mix, damn near every good becomes a rich bundle of attributes for which profit-maximizing suppliers are working diligently to not just meet the needs of their customers, but serve the terms of the explicit and implicit contracts from which any deviation brings the spector of margin-spoiling transaction costs. There’s a lot of gravity at the status quo. Which, in a way, is simple rediscovering menu costs, but with the important distinction that just because the number on the menu hasn’t changed doesn’t mean the price hasn’t. The menu is a lie.

So, yeah, measurement is hard.