Statistical Mirage

The Economic Research Service (ERS) is trumpeting its latest findings that consumers can buy seven servings of produce a day for about 64 cents — or could at least back in 1999 when the data the study was based on was collected.

Along with the study, the Produce for Better Health Foundation vows to run “extensive outreach to consumer media and nutrition educators to spread the news” and proclaims that the produce “…industry has just been handed a very powerful tool.”

Perhaps. But whenever consumers report something in a few focus groups — in this case that cost is a barrier to consuming more fruits and vegetables — and the data shows that this is not true, it is usually a mistaken response to assume that consumers are wrong and need to be instructed as to the reality of the situation.

Especially when it comes to retail pricing, consumers are remarkably astute. Changes in the price of beef can be tracked on sales of chicken. Announce zero percent financing and car sales increase. All this is no less true of fruits of vegetables. So before we launch a national campaign to tell consumers they are misguided, maybe we should think about ways in which this study may not tell the whole truth.

1) What items do people actually want to eat?

The study came up with its cheap estimate of costs by using the least expensive items in each category. But there was no attempt in the study to track whether those items actually mirror consumer market baskets. The study was an example of what highly motivated people could do if they so desired, but there is no evidence that people are so motivated.

Although the study trumpeted the idea that its produce selections would account for only 12 percent of daily food expenditures, merely flipping the chart and taking the choices from the most expensive items, rather than the least expensive, dramatically changes the results. In fact, using only the most expensive fresh fruits and vegetables studied means the seven servings of produce per day would cost not the 64 cents a day, but instead $4.44 a day, or over 80 percent of the $5.50 each person was spending on all food back in 1999. And the study excluded specialty produce, although items such as Bok Choy can be staples for certain types of cuisine.

Let’s try to be fair: If we just take the median of the fruits and the vegetables, separately, and figure three fruits and four vegetables, the price per day comes out to $1.48, or more than double the number headlined in the study and over a quarter of the per person food budget back in 1999.

If we take the mean or arithmetic average of the fruit and vegetable categories separately, our seven servings work out to $1.79 or about a third of the total food budget. No wonder consumers complain about the price. If they spend a third of the food budget on produce, that leaves a real squeeze on meat, dairy, deli, bakery, seafood and other items.

In effect, the study is like looking at meat and poultry prices and noting that by eating the cheapest cuts sold, one can meet one’s protein requirements cheaply. It may be true but it is a very defensive posture for an industry to take. Surely it makes more sense to promote the value of foods consumers want to eat.

2) Is the product available?

The whole study is based around the insight that the cost to purchase a produce item is not the same as the per serving cost — or what the study calls the “price to eat”. Sometimes the price to eat is higher because a portion of the item may not be eaten; usually it is much lower because the unit of sale is much larger than one serving.

Of course, this very fact, combined with the fact that the study doesn’t include fresh-cut items, means that the study is, quite literally, academic. It is impossible for a consumer to purchase individual servings of produce because they are not sold that way.

So the study may say that a serving of fresh avocado costs only 25 cents. But this is completely alien to a consumer’s real experience where the consumer might have to pay $1.99 for even one avocado.

3) Don’t forget the shrink.

The study adjusts for shrink caused by an intentional act, such as people not eating the rind of a watermelon. However, it assumes that other types of shrink stop at the retail store.

Consumers throw out lots of produce every day. Just as a store can’t price assuming zero shrink, consumers don’t experience the cost of fresh produce with zero shrink. They know they will wind up throwing out that one overripe banana, nobody will finish that head of red leaf lettuce, some moldy blueberries are in there, and sometimes the navels are dry and inedible.

If they can only consume 80 percent of the produce, then consumers’ experience of the cost of produce is going to be different than the study implies.

4) The poor have other problems.

In light of the small cost of food overall, it is doubtful that there are many affluent people or even middle-class people who are not eating produce because it is too expensive. So if this study were to be meaningful for anyone, it would have to be the poor. But poor people have many other problems.

First, many poor people live in inner cities where all food items cost more, including fresh produce. The data the study is drawn from does not account for the inner city poor who often buy their produce at local bodegas. So for all we know, the poor could actually be paying a lot more than this study asserts.

Beyond money, though, produce is an expensive purchase in terms of time and effort. Perhaps a poor woman might get together with a girlfriend, borrow someone’s car and go once a month to a Sam’s Club or true supermarket in the suburbs. She may stock up at good prices on canned goods and other needs. But she can’t stock up on perishables such as produce, which means additional trips to either the big store or more expensive local vendor.

It is like health care and the poor. You can give away free medical care and a lot of poor people still won’t get medical care when they need it. Price is just one barrier to entry. People have no transportation, can’t afford the time off from work, are afraid of the doctor, etc. Equally, you could give produce away for free and find that people have reasons for not eating it.

Like most research, the ERS study presents a lot of interesting data, but it would be a mistake for the industry to run to consumers with a wagging finger and an “I told you so” attitude. It makes more sense to think about addressing the way consumers actually experience buying produce than promoting a statistical mirage.