Find out how Covid-19 has impacted food production in your county

Issue 158: The Purdue Food and Agriculture Vulnerability Index has some data issues but is a fascinating and fine-grained look at local ag during the pandemic.

Hello!

This is Nosh Box, a lunchtime-ish food newsletter.

Read yesterday’s dispatch: The paradox of those "Immigrants Feed America" shirts

Here’s what I’ve got comin’ your way today: The Purdue Food and Agriculture Vulnerability Index, which was created by the Department of Agricultural Economics at Purdue University in collaboration with Microsoft.

The tool, launched about a week ago, is a web app that pulls from an enormous dataset to show the percentage of agricultural production lost due to farmworker illness. Here’s what the dashboard looks like:

(screengrab of the Purdue Food and Agriculture Vulnerability Index Dashboard)

You can view national and state data for production of beans, veggies, cattle/milk, wheat, and more. Plus, you can break down the data county-by-county, which can be accessed by right-clicking on the state in question. The tool aggregates Covid-19 data from Johns Hopkins University, Census population statistics, and USDA numbers on farm employment and production.

There’s SO much data, so you really have to worm your way through by using the right-click “Show As Table” function. This way, you can actually see the units they’re measuring, which is… helpful. (This took me an embarrassingly long time to figure out. Can’t win em all.) Just for fun — looks like wheat production in Hennepin County, Minnesota (the urban/suburban county I’m from) is down by 47.66 bushels this year; statewide, wheat has dropped by over 328,000 bushels or a whopping 0.42%. Small potatoes, so to speak.

The upshot of this whole data set is that U.S. ag production in seven major categories — beans, cattle, vegetables, hogs, chicken, wheat, rice — is down by less than 1% each across the board as a result of farmworker illness.

---

This tool is important but doesn’t tell the whole story…

…as Dan Nosowitz points out over at Modern Farmer:

The number of active workers isn’t necessarily the only problem facing food production right now; it’s a big one, but there are also major infrastructure challenges. Farmers dumping their crops are still able to produce, but they won’t show up in the food system due to a lack of buyers or a way to get that food to existing buyers.

Plus, I have some bones to pick with their data work:

For one, if you look at their methodology sheet (same URL as the dashboard; link at the bottom of the page), you’ll find out that the data points for agricultural workers with Covid-19 are expected values, not measured ones. They’re calculated on the assumption that ag workers contract Covid-19 at a rate equal to the rest of the population — which we know is not always the case.

And for two, because the index relies on farm employment data from the USDA, undocumented workers are almost certainly WAY undercounted. On top of that, Covid-19 case data is only as reliable as the state that’s collecting it, which is not always, uh, a reassurance of accuracy. What this speaks to, more than any fault of Purdue’s, is the need for non-political national data collection — funny, the USDA has an agency called the Economic Research Service that’s meant to do exactly this, but the Trump administration gutted it a few years ago. (And also maybe we need a political environment where tallying up people who’ve contracted a disease isn’t a partisan issue, but hmm, maybe that’s just me.)

But at the end of the day, I think the Agriculture Vulnerability Index is a fascinating tool that can help people get a better understanding of the local food system in their area, even if one that needs supplementation. And a benefit of huge data sets like this is that they can be combed for problem areas/loss hotspots in the food system, which could (hopefully!!) help target farm/farmworker relief efforts. It’s an admirable and ambitious effort from Purdue and ALSO speaks to the need for better agricultural research at a federal level. Check it out here.

(look at this stock photo of a stereotypical American farmer driving a tractor)

One last thing — an update on a topic we talked about in last week’s newsletter about why it’s so tough to get produce from farms to food banks. The USDA offered $3 billion in grants to buy produce from farmers and deliver it to those in need, and they awarded those grants to private-sector folks who may or may not have actually been qualified.

I mentioned that a lavish event planning company in Texas, called CRE8AD8, got $39 million to pack 750,000 boxes of vegetables, meat, and dairy, and deliver them to nonprofits from Arizona to Arkansas by June 30, per the San Antonio Express News. A massive mandate — and the Express News reports that, as of yesterday (May 26), no nonprofit had received a single box from CRE8AD8.

Now legislators are getting involved. Rep. Joaquín Castro and Rep. Lloyd Doggett, both Democrats from Texas, have each written a letter to USDA Sec. Sonny Perdue (no relation to the university mentioned above) calling for investigations.

Doggett did not mince words in calling for a complete stop-work order for CRE8AD8:

“A family cannot eat an IOU. Poor performing, inexperienced contractors risk delaying food delivery or even delivering spoiled, dangerous food to families who need help now.”

I’ll keep you posted on this story.

See you tomorrow!