In Defense of the Almond
Almond shaming.
That's what Californians called it back in 2015. According to a local news report, "it started when this statistic was released to the public: It takes one gallon of water to produce one almond."
The latest version of almond shaming is a May 13, 2023 New York Times opinion piece by Nicholas Kristof entitled "When one almond gulps 3.2 gallons of water".
To be fair, I have no idea whether Mr. Kristof bears a grudge against almonds. His concern is that in the drought-stricken Southwest, water is distributed irrationally and cheaply. If water were just sold at market rates, farmers would make smarter decisions about its use, and disasters such as mega-droughts and wildfires would be averted.
In Mr. Kristof's view, one of those smarter decisions would be to stop growing almonds, because the water would cost too much. Doing so would be enormously consequential, given that 80% of the world's almonds are produced in California.
The pause
If you're a regular reader, you know that I'll be talking about more than just almonds in this newsletter. Each week, I pause for a moment to describe how my topic illustrates the historically transformative power of statistics.
In this case, a single statistic ("It takes one gallon of water...") touched off an important, contentious debate about how we allocate one of our most precious resources. Legislators, policy experts, and the general public argued over a kind of statistic that only became possible in the 20th century, thanks to developments in the agricultural and statistical sciences.
Nearly a decade later, a revised version of that statistic ("it takes 3.2 gallons of water...) continues to figure in this debate. And yet, as I will show in this newsletter, the statistic is deeply misleading. Using that 3.2 gallon figure to imply that water is being wasted, or misused, is deceptive, because the number has been taken out of context.
This illustrates one of the most common ways our society has been "statisfied". Statistics that first became available in the 20th century can be uniquely informative, but when they get quoted out of context, what begins as information quickly becomes disinformative.
Why is this important?
Nicholas Kristof is not just a brilliant journalist. His work has had a uniquely powerful influence on the some of the industries and people he writes about. But before we start boycotting almond milk, changing water policies, and driving almond growers out of business, we need to understand that 3.2 gallon statistic.
I'm not a huge fan of almonds, by the way. Amaretto is nice, and marzipan brings back memories of the four years I spent in Germany as a child, but most almond products don't appeal to me. My "defense" of the almond is spurred by more academic concerns. I'd hate to see lives and industries disrupted by a statistic.
Agriculture is responsible for an estimated 92% of the global water footprint. Crudely speaking, it takes a lot of water to make food. This makes it easy to get caught up in one of those statistical factoid wars. (3.2 gallons of water to produce an almond? Sure, but it takes 53 gallons to produce one egg, and over 600 gallons to make a hamburger....) These statistics need context.
The water footprint
The total "water footprint" of a product consists of three kinds of water needed to produce it: Rainfall, managed water (e.g., irrigation), and the water polluted via its production.
The one-gallon-one-almond statistic came from studies that focused on managed water. The 3.2 gallon statistic comes from a more recent 2019 study that considered the entire water footprint.
Mr. Kristof referenced this study in his Times article this Saturday. Unfortunately, he seems to have lifted that 3.2 gallon figure from the first line of the Highlights section of the study (which is the first thing an online reader sees, even before getting to the abstract). The details of this study, and others, provide necessary context.
Nutrition matters
If you worry about that 3.2 gallon-per-almond figure, the first question to ask is: How does it compare to other foods?
Simple head-to-head comparisons tell us that, yes, almonds have a bigger water footprint than most crops, though not as big as meat and dairy products. But are head-to-head comparions sensible? Not really. A cup of iceberg lettuce, for instance, which requires very little water to produce, contains 2% or less of suggested daily values for fiber, protein, and a few nutrients. The nutritional content is close to zero. In contrast, a cup of almonds is a meal. Not a very balanced one, but you'd get approximately half the amount of protein you need in a day (short two amino acids), half the fiber, and over a fifth of the calcium, iron, and potassium.
In short, directly comparing water footprints isn't much use without considering nutritional value.
In the 2019 study that Mr. Kristof referenced, researchers at Cal State Sacramento and UC Davis calculated and ranked water footprints and nutritional values for 43 different crops grown in California. ("Nutritional value" was a composite measure based on content of protein, fiber, vitamins A, C and E, calcium, iron, magnesium, and potassium.)
One of the researchers' graphs, pasted in below, sums up the results quite nicely, though the graph is a little hard to decipher at first.
On the y-axis of the graph, lower values correspond to higher nutritional rankings. (This is counter-intuitive – usually, the higher you go on the vertical axis, the higher, stronger, better, faster, etc. the value. No matter. Just keep in mind that on this particular y-axis, lower is better.)
On the x-axis, the farther right you look, the greater the water footprint.
If you're wondering whether you're viewing the graph correctly, just look to the lower left and notice that spinach is a big winner. It's one of the most nutritious foods and it has one of the lowest water footprints.
Looking at the lower right, you can see that almonds have a higher water footprint than the other crops, but they're also one of the highest in nutritional value. That's the first point that can be made in their defense: There's a big nutritional payoff for the water they use.
What about almond milk?
Almond milk gets more attention than other almond products, given increasing consumer interest in nut milks (and pushback from the dairy industry).
An almond shamer who reads the previous section might say: fine, almonds are nutritious, but almond milk isn't, at least when you compare it to dairy milk. Why waste so much water on the almonds that go into making it?
Well, for one thing, generalizations about nutritional content don't make much sense given the diversity of almond and cow milk products currently on the market.
Cow's milk has several times the protein content of unfortified almond milk, for example, but you can actually buy almond milks that are fortified to have more protein than cow's milk. You can also buy cow's milk that's fortified with extra protein. So, who's the winner in this protein arms race? It depends. You have to read product labels.
Much the same can be said for other nutrients. Cow's milk is more nutritious than unfortified almond milk, but you can buy almond milks that outperform dairy on almost any nutrient of choice, if not all of them. Again, product labels tell the tale.
Cow's milk also contains substances that can be considered undesirable, but my goal here isn't to pick on it. Rather, I just want to emphasize that almond milk doesn't necessarily lag behind in nutrients. Meanwhile, studies conducted in 2017 and 2020 estimate that the production of dairy milk requires 1.8 times as much water as almond milk. (The 2020 study also estimates that the impact on global warming is 4.5 times greater for dairy milk than almond milk. Other studies confirm that almond production is similar to most crops in this respect but much less impactful than meat and dairy products.)
In sum, dairy milk is extremely nutritious, but almond milk can be quite nutritious too, and it requires less water to produce.
Economics matter
Almond shaming seems a bit unrealistic, given that the value of a crop tends to be related to its water footprint. Generally speaking, more valuable crops use more water. This is illustrated by another somewhat counter-intuitively designed graph from the 2019 study.
In this graph, the lower the value on the y-axis, the more valuable the crop. (Market values were averaged across the time period 2004-2015.) The farther right you look on the x-axis, the greater the water footprint.
The main thing to notice about this graph is the downward trend from upper left to lower right. There are outliers, such as strawberries and oats, but for the most part, crop values increase as their water footprint increases. As you can see, almonds are the most valuable crop, and they have the greatest water footprint.
(Below is the same graph without the outliers, which allows the downward trend to be seen more clearly.)
This graph illustrates one of the reasons almond shaming seems unrealistic. The problem, if you believe there is one, isn't almonds per se, but rather the correlation between a crop's value and its thirst. Asking the agricultural industry to produce less water-consumptive crops is tantamount to asking for less valuable crops.
The almond shamer might reply: Well, let's not pick on all the crops. Just almonds. Good luck with that. California exports nearly $5 billion worth to foreign countries alone; that's just over one-fifth of the state's agricultural exports. Simply complaining about the water footprint of almonds, without considering their nutritional and economic value, won't get you very far.
Bottom line
It's very, very catchy to say that it takes 3.2 gallons of water to produce an almond. Even now, as I write this section, I feel the pull of that statistic. It sounds like a lot of water.
At the same time, the statistic is deeply misleading. Almond production may require more water than other crops, but the results are highly nutritious and more lucrative. Even almond milks can be quite nutritious, and they require less water to produce than dairy milk.
What I just wrote doesn't seem very catchy, but we need more than sound bites. The American Southwest is contending with major water shortages, and Mr. Kristof is surely right that part of the problem is the way water is distributed there. But almond shaming is not a fruitful part of the solution. Simply forcing almond growers to pay more for water in hopes that they stop growing almonds – or persuading consumers to boycott almond products – would be needlessly disruptive to the industry and to the people involved.
Solutions
Complaining about almonds gulping water is unhelpful. Citing a statistic makes it worse, because statistics create a sense of legitimacy, and they may seem, falsely, to be understandable without context.
Legislative remedies are needed to better manage water distribution in the Southwestern states. As for almonds, what's needed, as with any crop in any state, is the refinement of strategies for reducing the environmental impacts of production.
California is already doing that. For instance, the Almond Board of California, a federally-mandated industry group that engages in compliance as well as promotion activities, reports that the amount of water needed to grow almonds was reduced by one-third between the 1990s and the 2010s, thanks to advances in micro-irrigation techniques. Meanwhile, one of the Board's sustainability goals is to create an additional 20% reduction by 2025.
I'm not able to evaluate progress toward that goal. I can only say that given the complex relationships between nutrition, economic value, and environmental impacts, reducing the water footprint of any crop seems like a worthwhile venture. Much more worthwhile than singling out any one crop on the basis of a statistic that misrepresents the broader story.
Thanks for reading!