What comes to mind? Probably words like fresh, local, community.
How about competition... collusion... even price? Didn't think so! So it is with the economists and other academics who have studied the modern farmers market phenomenon in the US. The more romantic, "civic agriculture" aspects of farmers markets dominate our conversation about them, as well as the research literature. Consequently, there's been surprisingly little effort put into understanding how the "markets" in "farmers markets" actually work.
To an economist, one critical aspect of how markets work, of course, is competition. And one way to assess the state of competition is: what happens to prices when more vendors are competing to sell a product? In my recently published Masters thesis, an original Food Industry Center dataset allowed us to look at precisely this dynamic across five different metro areas. The data also let us look at the interesting question of whether farmers market prices seem to have much to do with prices at nearby supermarkets.
I'll share the outcome in a minute... but why do we care? After all, aren't farmers market shoppers supposed to be folks with plenty of money, who really just come for fresh food, open air, a sense of community, and supporting local agriculture? Do prices and competition even matter?
Here's my argument, in a nutshell. Farmers markets are exploding - even right through the recession, they opened at record pace. More and more Americans are shopping there, of all types, and the old stereotype is fading (the paper traces this development, at some length, through the literature). In addition, the US government has explicit policy goals to encourage more markets in lower-income neighborhoods, and to expand the use of WIC and food stamps in farmers markets. So while price is certainly not the only thing that matters at the farmers market, it does matter - more all the time.
Now, for the good stuff...
So: is there much competition at the farmers market? Consider a couple examples. Here are the median and minimum recorded prices in the market for every week lettuce was on sale in 2009 at a farmers market we call TwinCities1. Each point represents a separate weekly observation.
The paper includes another chart showing that prices don't do anything terribly interesting when plotted chronologically (with date on the X-axis). But plot them by the number of vendors, as seen here, and bingo: the more people who show up to sell lettuce, the lower the price goes. A bigger lettuce market is associated with a better deal for consumers. Even in an "oligopolistic" market with imperfect competition (and farmers markets certainly qualify on a number of fronts) this is generally what theory would predict. And - always a nice side-benefit - it conforms with common sense.
Now take a look at another market: Fuji apples for sale in Syracuse. This graph also includes the maximum recorded price in the farmers market for each day.
Whoa... something's weird here. In this case, the minimum price goes down as vendors increase; as more people set up shop, at least one vendor chooses to respond by trying to underprice the other guys. But the median price prevailing in the market hardly budges; if anything, it increases a little.
The maximum prices show that at the same time as some vendors are lowering prices, others are raising them (or new, higher prices vendors are entering the market). Consumers shopping for apples here don't seem to benefit from "competition" at all (unless they are particularly intrepid, and make the effort to seek out the minimum-price vendors).
In all, the data allowed me to analyze 23 different "product-markets" like this, across five U.S. cities. Many unambiguously "worked" like the lettuce market above. A few were statistically ambiguous. And a number of others were "backwards," you might say, to a statistically significant degree (that is, prices actually went up as more vendors moved in).
Why competition some times, and not others?
One distinction appeared to be critical, and a meta-analysis of the full set confirms it is highly significant: perishable products (in this sample, lettuce and blueberries) are generally competitive; less perishable ones (apples and frozen ground beef), much less so.
One reasonable explanation for this is just common sense, but also fits like a textbook case of a basic, so-called "Cournot" market. The Cournot model presumes a closed market, with a limited number of competitors who choose a production quantity with which to go to market. Similarly, farmers market vendors must literally choose how much to put on the truck each day. Vendors of products that can spoil are compelled (as in Cournot) to sell what they've brought, and they drive the price down as the number of competitors grows.
But if you're selling hearty apples or ground beef... well, it's a lot easier for you to just throw your unsold inventory back on the truck for another day. The Cournot patterns doesn't apply to you... or at least, the incentives are much weaker.
Just what logic does apply, then? That's when this stuff gets really fun to think about! When competition's not happening, there must be other factors at play. How easy is it for customers to browse the market and compare prices? How easy is it for new vendors to enter the market and start selling there? If a product is easy to sample (like apples) what incentives does that create for people to shop around... or not; once you've sampled, do you feel a greater obligation to buy?. (Hat-tip to my adviser Rob King for bringing up his apple-sampling experience.)
Perhaps the most provocative possibility: are vendors fixing prices? The same sense of close-knit community and camaraderie that we might treasure about farmers markets can also be seen, through an economic lens, as anti-competitive collusion. There is evidence from the literature that vendors can indeed form cartels, checking with each other to keep prices at a certain level (and turning the cold shoulder to vendors who might try to undercut them). Indeed, where market prices seemed unusually static, I could look at the raw data and perceive instances where prices moved in sync to an uncanny degree, and rarely to the consumers' advantage. This is certainly not the rule, but it appears to happen.
All of these insights suggest possible policies that market managers or the government, where it has a stake, might explore to encourage greater competition and lower prices. Of course... is that what we want, if it spoils some of what otherwise makes farmers markets special? That's certainly not for me to answer!
Finally, there was the secondary issue I mentioned above: how well do farmers market prices track the prices in nearby supermarkets? According to this dataset, at least, the answer is: not well at all. In fact, there appear to be very few cases where prices in farmers markets and supermarkets move together to any significant degree. They are their own world, seemingly divorced from whatever forces set the prices in more conventional settings.
And how about one last question, which just about everyone I talked to about this research seemed to ask me: are farmers markets cheaper? Not according to this data, at least in any categorical sense. But nor are they categorically more expensive; local factors and product differences would appear to be critical here.
Feel free to leave your comments below, and happy shopping! For more, I invite you to dig into the paper. Either way, I hope you'll never look at farmers markets the same way again!