the money

Instacart and the On-Demand Middle Class

Inside A Whole Foods Market Inc. Store As Profit Tops Estimates
Whole Foods is one of Instacart’s partners. Photo: Ty Wright/Bloomberg Finance LP/Getty Images

It is the middle of the afternoon and I am standing in a crowded Whole Foods trying to read another person’s mind through an iPhone. When he asks for “organic” eggs, does he really want “humane”? Is “kosher for Passover” the same thing as “kosher”? Does he want a mix of ripe and green bananas? (When I buy my own bananas, I get half green, half ripe. Everybody should do that.)

Earlier this year, I spent some time moonlighting as an Instacart shopper, picking out and hand-delivering groceries across Washington. There’s a saying that often gets applied to investing, dieting, and marathons: It’s simple, but not easy. Grocery shopping, on the other hand, is the opposite phenomenon: easy, but not simple. There is a surprising amount of judgment that goes into filling up a cart, especially when you are not making decisions on your own behalf. You get the wrong eggs. You take too long picking out bananas. You grab the wrong items, full stop. The customer ends up annoyed or disappointed. The order ends up late.

Of late, Instacart has found a radical way to improve its workers’ decision-making: hiring them. The company signed me up as an independent contractor, making a bit of money every time I got assigned an order and fulfilled it. But earlier this year, it started offering contractors in certain markets the chance to become a part-time employee instead. And today, it announced that contractors in all 16 cities where the company operates, including in New York, can apply to become part-timers. “This has been a great change for Instacart,” Apoorva Mehta, the company’s founder, said in a press release. “Those that are now employees report a new feeling of connectedness to Instacart.”

Instacart is not the only start-up bucking the prevailing “1099” or “gig economy” trend, wherein companies act as matchmakers between customers and scads of thinly compensated contractors. Sure, the biggest and most famous of these companies still and will probably always use contractors, among them Lyft and Uber. The work there is more straightforward and less personal, and the companies’ business models are predicated on having little labor overhead. But Instacart is hardly alone in recognizing the benefits of old-fashioned employees. A number of start-ups, including Alfred and Shyp, have, too. Taken together, they are showing that the growth of the on-demand economy does not mean the growth of exclusively low-pay, low-skill jobs. There just might be room for an on-demand middle class.

To see why parts of the gig economy might function better without the “gig” part, my experience is probably instructive. Getting the job with Instacart was simple. I filled out an online form and took a test where I identified different food items. After that, I met with a manager who gave me some quick advice about the quirks of the business. Check the eggs. Get the coldest milk you can. Don’t buy anything you would not eat, and for goodness’ sake don’t squish the bread. Spend some time reading your customer, too, she advised. “A lot of people want a call about their groceries, just to feel special,” she told me. “They’ve had a long day. They just want someone to say, ‘Is there anything that you forgot? Do you want that avocado ripe or not?’”

After that quick session, I was hired and went to work. It is a surprisingly fun side-gig for a professional nosy nancy, which is pretty much what journalists are. There was the rush to get to the customer who wanted a pregnancy test, crackers, and dish detergent left wordlessly on her porch, a grocery order seemingly written by Richard Yates. Another time, I arrived outside a familiar-looking house in tony upper Northwest. It belonged to a former Obama administration official whose dinner parties I had attended some time ago. Could I ditch the bags, push the doorbell, and run? Thankfully, a babysitter answered the door.

Fun aside, I was not particularly good at the job. I was slow to find items in-store. I imagine I got some things wrong — something that has annoyed me when Instacart has brought me my groceries, and is a major complaint among friends of mine who use the service. When you are paying to get flat parsley delivered instead of cilantro, or overripe bananas, or liquified ice cream, what’s the point? I might have also missed some items, another major complaint, given that it might necessitate a trip to the store even after spending some money to avoid that hassle.

Making contractors into employees helps to mitigate those issues, Instacart has found. It is able to give employees better training and supervision, improving their skill at the job and leading to happier shoppers and customers. “We found a lot of improvements in efficiency,” Andrea Saul, the company’s communications director, told me. “The shoppers got better and more accurate, they were faster, and we had more on-time deliveries.” She added, “Though this has caused us to incur greater costs, it is better for the business.” (It is also worth noting that hiring employees rather than using contractors might significantly reduce the company’s legal liabilities.)

Granted, converting contractors to employees comes with a price: Instacart contributes to Social Security, Medicare, unemployment insurance, and workers’ compensation on behalf of its new employees, and it pays them above the local minimum wage. But some of that cost might be offset by lower rates of employee turnover, as well as better sales. When I was a shopper, I made about $60 a shift, from which I needed to pay for gas, car insurance, my phone, my data plan, and federal and local taxes. All in all, my guess is that it netted out to a little less than the minimum wage. That’s not really a gig worth keeping for long. But a steady part-time job with some benefits and a decent-enough per-hour rate? For a lot of workers, it might be.

This is all straightforward Econ 101. Hiring workers gives a company more control over them. It improves the quality of service. It helps the business fulfill tasks requiring some experience and judgment. It reduces labor turnover. As such, W2 employees make a lot of sense for more-complicated on-demand economy businesses, like Instacart. On top of that, it shows that the app economy can create decent, flexible, stable, and even middle-wage jobs, ones that require some finesse and expertise — not just ones where coglike contract workers make less than the minimum wage. At least until the robots get better at picking out quinoa.