Determining Poverty

This morning I was confronted by one of the more depressing Substacks I have read lately and wanted to comment on it. After a preamble Michael W. Green comes out strong:

And so now, let’s tug on that loose thread… I’m sure many of my left-leaning readers will say, “This is obvious, we have been talking about it for YEARS!” Yes, many of you have; but you were using language of emotion (“Pay a living wage!”) rather than showing the math. My bad for not paying closer attention; your bad for not showing your work or coming up with workable solutions. Let’s rectify it rather than cast blame.

I have spent my career distrusting the obvious.

Markets, liquidity, factor models—none of these ever felt self-evident to me. Markets are mechanisms of price clearing. Mechanisms have parameters. Parameters distort outcomes. This is the lens through which I learned to see everything: find the parameter, find the distortion, find the opportunity.

But there was one number I had somehow never interrogated. One number that I simply accepted, the way a child accepts gravity.

The poverty line.

I don’t know why. It seemed apolitical, an actuarial fact calculated by serious people in government offices. A line someone else drew decades ago that we use to define who is “poor,” who is “middle class,” and who deserves help. It was infrastructure—invisible, unquestioned, foundational.

This week, while trying to understand why the American middle class feels poorer each year despite healthy GDP growth and low unemployment, I came across a sentence buried in a research paper:

“The U.S. poverty line is calculated as three times the cost of a minimum food diet in 1963, adjusted for inflation.”

I read it again. Three times the minimum food budget.

I felt sick.

In the balance of the post Mr. Green analyzes that and arrives at the conclusion that the actual threshold household income below which people are poor is $140,000. Let that sink in for a while. The implications are staggering. By that reckoning not only are most Americans poor but there is no practical, foreseeable strategy for changing it. Contrary to his own criticism in that early passage he presents none, no doubt for that reason.

He introduces some worthwhile ideas. One of them is the “cost of participation”:

To function in 1955 society—to have a job, call a doctor, and be a citizen—you needed a telephone line. That “Participation Ticket” cost $5 a month.

Adjusted for standard inflation, that $5 should be $58 today.

But you cannot run a household in 2024 on a $58 landline. To function today—to factor authenticate your bank account, to answer work emails, to check your child’s school portal (which is now digital-only)—you need a smartphone plan and home broadband.

The cost of that “Participation Ticket” for a family of four is not $58. It’s $200 a month.

The economists say, “But look at the computing power you get!”

I say, “Look at the computing power I need!”

The utility I’m buying is “connection to the economy.” The price of that utility didn’t just keep pace with inflation; it tripled relative to it.

In 1955 (his benchmark year) about 75% of Americans had telephones in their homes so IMO that’s a reasonable factor in the “cost of connection”. I’m not entirely confident in his estimate of the cost of that—the canceled checks I have reflect a figure somewhat higher than that.

A lot more than the things he lists have changed since 1955. For one thing where people live has changed enormously since then. In 1955 35% of the population lived in the South. Today nearly 40% do and most of those live in the major cities of the South, e.g. Dallas, Houston, Miami, etc. The implications of that on his estimate of the change in housing costs is considerable.

Here’s a mind boggling figure:

Healthcare: In 1955, Blue Cross family coverage was roughly $10/month ($115 in today’s dollars). Today, the average family premium is over $1,600/month. That’s 14x inflation.

I think he’s underestimating the cost of healthcare insurance in 1955. I have the check stubs to prove it. I do wonder what the total effect that monstrous growth in costs has on the overall cost-of-living figure he’s coming up with when you consider primary and secondary effects.

I also think that he’s making some bad assumptions. For example, two jobs in a household does not require two automobiles (and maintenance, insurance, etc.) It doesn’t even require one. One or both can take public transportation. For the first several years of our marriage my wife and I were both employed but shared a single car. My understanding is that a lot of people who live in New York City, for example, don’t own cars at all.

Furthermore, averages are not as meaningful when analyzing things that don’t occur in a standard distribution than than they are in things that are. The top-selling smartphone in the country is the iPhone Pro Max ($1,299). The tenth largest seller is the moto g ($159) . Either achieves the connectivity about which Mr. Green speaks. An iPhone is a luxury (and maybe a status symbol) not a necessity.

I recommend reading the whole thing. Despite its flaws, I think he’s put his finger on something, namely the factors behind the widespread insecurity. People with household incomes between the official poverty rate but less than $100,000 have a lot to be insecure about.

And it explains the widespread insecurity about illegal immigration.

Another thing it highlights is how problematic the federal approach to making policy is. Several differing conclusions might be drawn from that. I think that more should be done at the state level and less at the federal level (sadly, that doesn’t help Illinois or Chicago much).