How do we fight the intractable, devastating public health problem that is obesity?

It’s a conversation I’ve had many, many times, and invariably someone suggests that we have to fight obesity by fighting poverty. Increase the minimum wage, ensure decent housing, provide medical care. If we lift people out of poverty, the obesity problem will ease.

It’s an idea invariably put forth by people who care deeply about both issues, but it has a problem: The evidence doesn’t support it. In the United States, the poorest are not the fattest; that dubious distinction goes to the middle-income tier.

The difference isn’t large. If we look at people with income below 130 percent of the federal poverty level (which translates to $32,630 for a family of four), 39 percent of them are obese, vs. 41 percent for people with income between 130 percent and 350 percent of poverty level ($32,630 to $87,850). Cynthia Ogden, of the Centers for Disease Control and Prevention, warns against reading too much into that two-point difference — we can look at the two groups as just about equally fat — and points to more pronounced differences when we look at specific groups.

Take women. Among Hispanic women, the lowest-income third does indeed have the highest obesity rate (48.7 percent). Among white women, the low and middle tiers essentially tie (at 42 percent and 42.5 percent, respectively). Among black women, who have the highest obesity rates overall, the low-income tier also has the lowest obesity rate (55.8 percent).

Men look quite different. Overall, the low-income tier has the lowest obesity rate (31.5 percent) — and that’s true of both white and black subgroups.

I could go on. But better that you do it, by visiting the CDC’s compilation of the data. I’ll just repeat that the poorest Americans do not have the highest obesity rates. It’s not true of adults, and it’s not true of children.

This doesn’t mean obesity is unrelated to class. There most definitely is a class element, but it’s not the poor vs. the rest. It’s the poor and the middle vs. the top.

According to Ogden, the CDC’s income classifications break the population into approximate thirds. With the exception of black men, the wealthiest third is never the heaviest. But when we move on to education, the two-thirds/one-third divide is even clearer. The obesity rate of college graduates is 27.8 percent, and the same for men and women, compared to just about 40 percent for the rest of the population.

College graduates are about a third of the population. Not exactly the same third as the high-income third, of course, but there’s a strong correlation. In a rough way, I think we can say that there’s about a third of the population that manages to navigate our food environment a bit better than the other two-thirds.

As much as I’d like to see us, as a society, fight both poverty and obesity, the idea that we can somehow boost everyone into the socio-economic top third is a Lake Wobegonian fantasy where we’re all way above average. Besides, even if we could, it’s not like the top third is a smashing public health success. A 30ish percent obesity rate looks good only when you compare it to the 40ish percent of the rest of the country.

And another besides — while we see the correlation between obesity and class, we don’t know whether it’s class that’s causing the lower obesity rates. “All of the work that I’ve seen is all associations. It’s not causal work,” says Ogden. All you can do is hypothesize, because “you can’t set up a controlled trial.”

And hypothesizing happens. Many of the hypothesizers look to data on people who get SNAP benefits (Supplemental Nutrition Assistance Program, formerly known as food stamps), versus those who qualify for them but don’t collect them, because those benefits are, essentially, an income boost. Craig Gundersen, an economist at the University of Illinois who studies SNAP, says he believes it’s “generally the case” that, if you gave poor people more money, they’d be less likely to be obese, and cites data that show SNAP recipients have “the same, or lower, rates of obesity” compared to SNAP-eligible people who don’t get the benefit.

But it turns out that experts disagree — vehemently — about this, with some contending that SNAP recipients have worse diets and higher obesity rates. It all depends on how you slice the data. It’s hard to parse an academic fight over data analysis and pull out a ringing endorsement for the theory that poverty drives obesity — or doesn’t.

It’s certainly possible that having more education and income arms you to better navigate our food environment. It’s also possible that being obese makes you downwardly mobile, and the causality works the other way. Or maybe a little bit of both. But, when the poorest aren’t the most likely to be overweight, it’s clear that there are plenty of other factors at play. I think we have to step away from the hypothesis that poverty is a primary driver of obesity.

We need a new hypothesis (besides the one about food access, which I’ll tackle next month). And, wouldn’t you know it, I’ve got one! It’s about as straightforward and obvious as obesity hypotheses get: Humans are simply ill-equipped to deal with a landscape of cheap, convenient, calorie-dense foods that have been specifically engineered to be irresistible. The inability to navigate our food environment is as near-universal as inabilities get.

Focusing on that inability, though, seems like blaming the victim. The last thing people dealing with obesity — let alone the one-two punch of poverty and obesity — need is to be told is that it’s their fault. But think about it this way. The people engineering the food that’s all around us are doing it with the specific intention to override ordinary human willpower. (Read Michael Moss’s excellent book, “Salt Sugar Fat: How the Food Giants Hooked Us,” on that subject.) They’ve been spending billions of dollars on it. Is it any surprise that they succeed?

In what may be a first for this column, I have a sports analogy. Let’s say you play Serena Williams at tennis. You are, of course, trounced. Do you blame your forehand? At some level, sure. But the real problem is that you have absolutely no business going up against the best in the world. Those food engineers with the billions of dollars? They’re Serena Williams. We’re just us, and our willpower is our forehand.

The food environment has overpowered the lowest socio-economic tier. It has overpowered the middle socio-economic tier. It has overpowered, to a slightly smaller degree, the top socio-economic tier. The food environment, not poverty, is the culprit.

More from Food: